BIOARCHAEOLOGY OF VIOLENCE AND SITE ABANDONMENT
AT CASAS GRANDES, CHIHUAHUA, IvIEXICO
by
CHRISTOPHER MICHAEL CASSERINO
A DISSERTATION
Presented to the Department of Anthropology
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Doctor of Philosophy
June 2009
11
University of Oregon Graduate School
Confirmation of Approval and Acceptance of Dissertation prepared by:
Christopher Casserino
Title:
"Bioarchaeology of Violence and Site Abandonment at Casas Grandes, Chihuahua, Mexico"
This dissertation has been accepted and approved in partial fulfillment of the requirements for
the Doctor of Philosophy degree in the Department of Anthropology by:
John Lukacs, Chairperson, Anthropology
Guy Tasa, Member, Anthropology
Frances White, Member, Anthropology
John Orbell, Outside Member, Political Science
and Richard Linton, Vice President for Research and Graduate Studies/Dean of the Graduate
School for the University of Oregon.
June 13,2009
Original approval signatures are on file with the Graduate School and the University of Oregon
Libraries.
---------------
© 2009 Christopher Michael Casserino
III
An Abstract of the Dissertation of
Christopher Michael Casserino for the degree of
in the Department of Anthropology to be taken
IV
Doctor ofPhilosophy
June 2009
Title: BIOARCHAEOLOGY OF VIOLENCE AND SITE ABANDONMENT AT
CASAS GRANDES, CHIHUAHUA, MEXICO
Approved: ~ _
Dr. John R. Lukacs
The objective ofthis dissertation is to address violence at the archaeological site
of Casas Grandes (Paquime) in northwest Chihuahua, Mexico. The reasons for the
abandonment ofPaquime are uncertain. The prevailing theory claims this geographic
area endured centuries ofwarfare, ritual sacrifice, and at least one massacre; this theory is
supported by numerous unburied bodies recovered at the site. These assertions of
violence have never been corroborated by osteological data.
Data were collected from a sample ofMedio period (A.D. 1200-1450) human
skeletal remains recovered from the 1958-1961 excavations at Casas Grandes. These
data were synthesized with accelerator mass spectrometry radiocarbon dates, fluoride ion
dates, population demographics, and burial context. Frequencies of ante-, peri-, and
postmortem trauma were compared to other studies from the Old and New Worlds.
vI argue that warfare was not endemic to this region and that a massacre did not
occur. Moreover, cannibalism and probably human sacrifice were practiced. I assert that
these activities may have been related to the proliferation of the Mesoamerican ballgame
in the American Southwest and to Paquime's role as the distribution center of the
region's ritual and exotic goods. This dissertation underscores the importance of
including skeletal analysis with other lines of archaeological inquiry when answering
questions about human behavior.
Vi
CURRICULUM VITAE
NAME OF AUTHOR: Christopher Michael Casserino
PLACE OF BIRTH: Anaheim, California
DATE OF BIRTH: January, 26, 1967
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon .
University of Montana
Eastern Washington University
Spokane Falls Community College
DEGREES AWARDED:
Doctor of Philosophy in Anthropology, 2009, University of Oregon
Master of Arts in Anthropology, 2001, University of Montana
Bachelor of Arts in Anthropology, 1997, Eastern Washington University
Bachelor of Science in Biology, 1997, Eastern Washington University
Associate ofArts in General Studies, 1994, Spokane Falls Community College
AREAS OF SPECIAL INTEREST:
Human osteology
Forensic pathology
Prehistory ofNorthwest Mexico
Forensic Archaeology
Museum collections management and preservation
PROFESSIONAL EXPERIENCE:
Anatomy and Exhibit Installation Specialist, Premier Exhibitions, 2008-present
Museum Collection Specialist, Northwest Museum of Arts and Culture, Spokane,
Washington,2004-present
Vll
Adjunct Lecturer, Eastern Washington University Departments of Justice Studies,
and Geography and Anthropology, Cheney, Washington, 2000-2008
Deputy Medical Investigator, Spokane County Medical Examiner's Office,
Spokane, Washington, 2000-2008
Graduate Teaching Fellow, University of Oregon Department of Anthropology,
Eugene, Oregon, 2001-2003
Graduate Teaching Assistant, University ofMontana Department of
Anthropology, Missoula, Montana, 1999
GRANTS, AWARDS AND HONORS:
University of Oregon Institute for Cognitive and Decision Sciences
research grant, 2008
Foundation for the Advancement of Mesoamerican Studies, Inc.
research grant, 2008
University of Oregon Graduate School research grant, 2007
University of Oregon Department of Anthropology graduate student research
award, 2002, 2007
Eastern Washington University Alumni Foundation grant, 1997
Dean's List, Eastern Washington University, 1995-1997
Lambda Alpha Honor Society of Anthropology, Lifetime
Hewlett-Packard Employees Scholarship, 1985
Vlll
ACKNOWLEDGMENTS
This dissertation was supplemented by emotional and financial support of many
people, institutions and foundations. First, I wish to thank professors John Lukacs, Guy
Tasa, Frances White, and John Orbell for their support and critique while doing the
fieldwork and writing this dissertation. Particularly, Dr. Lukacs has read and commented
on earlier drafts of this paper and others, and I feel the finished product is much stronger
because of his efforts.
Second, I am indebted to my wife, Alicia, and my children Christina and
Anthony, who have suffered years of displacement and neglect due to my pursuit of
higher education. They were my lifeline in times of extreme mental stress. My extended
family, both consanguineal and affinal, have also shown tremendous support.
Third, a number of friends and colleagues have contributed to the completion of
this dissertation in ways too numerous to mention. In the U.S., thank you to: Cam
Walker, Sarah Keller (Eastern Washington University), Michael Whalen (University,of
Tulsa), Todd Pitezel (University of Arizona), Sophie Kohn (University ofNew Mexico),
John Ross (Eastern Washington University, Emeritus), Art MacWilliams, John Roney,
colleagues at the Northwest Museum of Arts and Culture, and colleagues (past and
present) at the Spokane County Medical Examiner's Office. In Mexico, gracias to: Luis
Tena, Eduardo Gamboa, and Elsa Rodriguez at !NAB, Chihuahua; Roberto Garcia Moll
at the Consejo de Arqueologia, Mexico, D.F.; Laura Vasquez at the Museo de las
Culturas del Norte; Spencer and Emi MacCallum, and; the people of Casas Grandes. No
IX
gracias to the Mexican aduana agent who extorted fifty dollars from me while crossing
the border during the second trip.
This research was supported by grants from the Foundation for the Advancement
of Mesoamerican Studies, Inc. (Grant #07009), the University of Oregon Graduate
School, the University of Oregon Department of Anthropology, and the University of
Oregon Institute for Cognitive and Decision Sciences.
For Alicia, Christina, and Anthony
x
Xl
TABLE OF CONTENTS
Chapter Page
I. INTRODUCTION 1
Research Objectives 6
Hypotheses 8
Significance...................... 9
Organization of the Dissertation 9
II. THE CASAS GRANDES REGION 11
Defining the Geographic Study Area.................................................................... 11
The Site and Its History 13
Site Chronology ,. 18
Cultural Connections............. 19
Paquime's Genesis and Decadence 23
Mesoamerican Influence . 24
Development and Emergence ofPaquime 27
III. REVIEW OF THE OSTEOLOGICAL LITERATURE 32
Burials 32
The Nature and Properties ofBone....................................................................... 38
The Timing and Types of Bone Fracture 39
Cause and Manner ofDeath from the Skeleton.................................................... 41
XlI
Chapter Page
Case Studies.......................................................................................................... 46
Cannibalism Diagnosis 49
IV. MATERIALS AND METHODS.......................................................................... 53
Sampling............................................................................................................... 53
Data Collection 55
Burial Provenience Data 56
Sex and Age Determination 56
Methods of Data Analysis..................................................................................... 62
Dating.................................................................................................................... 63
V. RESULTS 66
Sample Population Demographics............................................ 66
Non-Interred Skeleton Burial Provenience 69
Antemortem and Perimortem Fractures 69
Peri-/Postmortem Fractures................................................................................... 72
Spatial Distribution 81
Skeletal Element Representation 82
Fluorine Dates 85
AMS Radiocarbon Dates 87
VI. EVALUAnON OF HYPOTHESES 89
Hypothesis 1: Intergroup Violence Was Endemic in the Casas Grandes
Region............................................................................................................. 90
Xlll
Chapter Page
Hypothesis 2: The Final Population ofPaquime Was Massacred 92
Hypothesis 3: Human Sacrifice and Cannibalism Were Practiced at Paquime 94
VII. DISCUSSION...................................................................................................... 98
Warfare.................................................................................................................. 98
Massacre.. 100
Human Sacrifice and Cannibalism. 102
Element Representation ofFragmented Skeletons........... 106
Non-Osteological Evidence: Chronometric Dating and Burned Structures 107
VIII. CONCLUSION 110
Contributions of This Research...... 111
Future Research Directions... 112
APPENDICES ;.. 113
A. INDIVIDUALS INCLUDED IN STUDY SAMPLE 113
B. DETAILS OF FLUORINE ANALySIS 116
BIBLIOGRAPHY 118
XIV
LIST OF FIGURES
1. Artist's depiction of the massacre at Paquime 2
2. Las Cuarenta Casas site Near Madera, Chihuahua, Mexico 4
3. Site excavation photo of "Burial" 14-1 5
4. Casas Grandes site plan......................................................................................... 17
5. Relationship of Casas Grandes with other chronologies.... ..... ..... ...... ... .... ..... ...... 20
6. Casas Grandes burial types 34
7. Demographics of the study population 67
8. Age distribution ofNIS and IS 67
9. Normal ulna, ulna with healed fracture................................................................. 72
10. Old adult male with healed cranial and facial fractures........................................ 73
11. Portion of child's parietal with hinge fracture...................................................... 73
12. Portion of frontal bone showing unhealed depressed fracture.............................. 74
13. Age distribution of group with postmortem modification 74
14. Longbone fragment showing various types of peri-/postmortem damage............ 75
15. Pelvic innominates ofa single individual showing postmortem removal
of iliac blades 76
16. Second cervical vertebra with cutmarks on the odontoid process 78
17. Longbone assemblage from a single individuaL.......... 79
xv
Figure Page
18. Partial skull showing multiple diagonal cutmarks running upper right to
lower left across the frontal bone 79
19. Charred and fragmented skull.......... 80
20. Distribution of postmortem modified remains by burial provenience and unit.... 82
21. Bone fragment count (arms and legs considered separately)................................ 84
22. Bone fragment count (alllongbones pooled)........................................................ 84
23. Fluoride ion concentration distribution................................................................. 86
24. Comparison of the distribution of fluoride concentration between populations... 86
25. Raw AMS dates plotted against their calibrated dates.......................................... 88
26. Comparison of antemortem and perimortem trauma between Paquime
and other selected sites.......................................................................................... 91
27. Complex burial 44 A-L-13.................................................................................... 105
XVI
LIST OF TABLES
Table Page
1. Skeletal sexing attributes 59
2. Total counts and frequencies of sex by sub-population........................................ 68
3. Provenience and count ofNIS skeletal material................................................... 70
4. Detail of skeletal trauma separated by timing....................................................... 71
5. Tally of fractures by timing and location.............................................................. 71
6. Summary of postmortem skeletal modification...... 77
7. Comparison ofPaquime with Southwest sites showing evidence
of cannibalism........... 80
8. Comparison of skeletal element frequencies between a laboratory skeleton
and Paquime............. 85
9. Summary ofAMS and fluoride data...... 87
10. Comparison of skeletal damage and preservation associated with
different types of corpse processing.. 89
XVll
LIST OF MAPS
Map Page
1. Location of Casas Grandes (Paquime) 14
CHAPTER I
INTRODUCTION
"Archaeological interpretations ofconflict rely on the preservation and
interpretation ofartifactual and human remains. The question arises:
Where are all the dead people or other material indicators that could
provide evidence ofthese wars for Nlesoamerican archaeologists? What
do physical anthropologists have to say about this matter?"
---M Nicolas Caretta (2008:387)
Casas Grandes, or Paquime (CHIH:D:9: 1) in northwest Mexico has been called
the largest and possibly most complex prehistoric community north of Mesoamerica
(Whalen and Minnis 2003 :315). In spite of this, the amount of research dedicated to the
Casas Grandes region is vastly overshadowed by that undertaken in the United States
Southwest and central Mexico. Thus, Casas Grandes has largely been overlooked for its
potential as a window into the connections between the two (Cordell 1984).
Many questions about Paquime remain unanswered. For instance, its rapid
development and abandonment remain poorly understood. While several scenarios have
been proposed for its rapid population growth, only one has been advanced for its
abandonment. Archaeologist Charles DiPeso theorized the city fell into economic
despair (DiPeso 1974; DiPeso et al. 1974). Then the people who, for "two and a half
generations sat idly by and watched the magnificent city ofPaquime fall into disrepair",
were massacred by invaders and the city burned (DiPeso 1974:2:319) (Figure 1). This
1
--- --------------
----------------------
2
theory, based on speculation and a discredited dating sequence, has never been subjected
to testing.
The general objective of this study is to use bioarchaeology to address conflict in
Northwest Mexico to give a clearer understanding of cultural dynamics. Larsen
(2002: 119) defines bioarchaeology as the study of human remains from archaeological
contexts. DiPeso primarily uses a toppled shrine and over 100 non-interred skeletons as
evidence of a massacre that followed centuries of regional warfare (1974:2:612-613).
LeBlanc (1999:252) believes that if this number of dead is extrapolated to account for all
those killed site-wide, the number could equal 1,000 to 2,000. This implies that one of
the greatest massacres in the prehistoric New World occurred at Paquime.
Figure 1. Artist's depiction of the massacre at Paquime (DiPeso 1974).
3LeBlanc (1999:200-203) argues that settlement patterns and locations changed
during the Hohokam Late Period (this corresponds to the Medio Period at Casas Grandes).
People aggregated in larger settlements that were built in defensive locations (situated
high or in relatively inaccessible locations) (refer to Figure 2, for example). Large sites
were often built around a good domestic water supply and had cisterns or reservoirs for
water storage. Architecturally, these sites could have bastion comers that facilitated
defense or could be fortified around their perimeter by the high exterior walls of the
apartment blocks, leaving a protected central plaza for public activities. Casas Grandes
shows some features that may have served defensive purposes (LeBlanc 1999:253). The
House of the Serpent (unit 11) has a bastion corner and other angled exterior walls that
could have facilitated defense of the structure. A related site, the Cerro de Moctezuma,
located 7 kilometers to the west, is a hilltop sentinel with a pueblo settlement located at
its base. Though its function is currently unknown, one possibility is its use as a
signaling tower (DiPeso 1974:2:362-364; LeBlanc 1999:253; Swanson 2003). In fact,
Swanson (2003) found that signaling 'stations' up to 75 kilometers away were possibly
integrated through a signaling system. Whether this system was for simple
communication or warning is unclear at present.
Lambert (1997) asserts that skeletal remains can provide a valuable source of
information regarding patterns of conflict in prehistory. She believes defensive sites and
structures can only suggest the threat of intrusion, while skeletal injury documents actual
events. Similarly, Caretta states: "Archaeological interpretations of conflict rely on the
preservation and interpretation of artifactual and human remains" (2008:387).
4Figure 2. Las Cuarenta Casas site near Madera, Chihuahua, Mexico is built into a
rock shelter high on the side of a river canyon. It is only accessible from below.
Thus, it is surprising that skeletal remains have been excluded from discussions of
violence at Casas Grandes in the fifty years since its excavation. The effects of warfare
and site abandonment were likely far-reaching due to its recognized importance as a
regional trading and ceremonial center in prehistory.
Redfern (2008:282) asserts that disarticulated or fragmentary remains have been
incorrectly perceived by archaeologists and biological anthropologists as having little
data potential. This is especially true ofPaquime as the fragmentary non-interred
skeletons have been largely excluded from analyses in the 35 years since DiPeso declared
them the remains ofthe massacre. Since then, studies of another non-interred assemblage
from Ram Mesa, New Mexico (Ogilvie and Hilton 2000) found evidence ofwitch killing,
while studies of similar assemblages from throughout the Southwest and Great Basin
found evidence ofcannibalism (Edgar and Sciulli 2006; Flinn et al. 1976; Haverkort and
Lubell 1999; Hurlbut 2000; Lambert et al. 2000; Novak and Kollmann 2000; Turner and
Turner 1999; White 1992). Similar studies from Mexico have also discovered
cannibalism (Caceres et al. 2007; Pijoan et al.2007; Villa et al. 1986). Evidence of
warfare and massacre has been detected on non-interred skeletal remains from the Great
Plains (Owsley et al. 1977; Willey 1990). This new era of osteological research opens
doors to the available interpretive possibilities.
Figure 3. Site excavation photo of "burial" 14-1, in room fill 20 em above living
floor. This is one of the deposits thought to be the remains of a victim of the
Paquime massacre. (Photograph number CG-F/61. Courtesy of the Amerind
Foundation, Inc., Dragoon, Arizona. Alfred Cohn, photographer.)
5
6Research Objectives
This dissertation will utilize multiple lines of evidence to answer three important
yet unresolved questions. First, was there violent inter-group conflict in and around
Casas Grandes, and if so, did it playa significant role in the abandonment of the city?
Second, did a massacre mark the end of the city's occupation? DiPeso's (1974;
DiPeso et al. 1974) massacre theory is based on burned structures and the recovery of
skeletons, which lacked evidence of formal burial. He believed the unburied skeletons
were the remains of the city's final occupants who were murdered and covered by
burning architectural debris as the city collapsed following ignition (DiPeso 1974, DiPeso
et al. 1974). These two objectives are integrally related since the second is largely
predicated on the first.
A third question will also be addressed. Were human sacrifice and cannibalism
practiced at Paquime? Human sacrifice in the New World is well-known (Benson and
Boone 1984; Gaither et al. 2008; Pijoan and Mansilla 1990). DiPeso writes oftrophy
skulls and skeletons that were found in what may have been sacrificial poses and
proveniences (under the ballcourts and posed around the base of a structural support
pillar). Did the Paquimeans practice other rituals involving destruction of the human
body? Although controversial, many osteological studies have documented the practice
ofcannibalism in the Southwest (Flinn et al. 1976; Haverkort and Lubell 1999; Hurlbut
2000; Novak and Kollmann 2000; Turner and Turner 1999; White 1992). Similar
accounts have been offered from Mexico (Pijoan et al. 2007).
7This dissertation will question many of DiPeso's interpretations of the
archaeological material. He advanced the idea that Toltec pochteca (merchants) traveled
to northwest Mexico and inspired the local inhabitants to build the city ofPaquime in
A.D. 1060. He believed the pochteca were responsible for the coordination ofthe city as
a regional production and trade center, sending exotic birds and other luxury items to the
Southwest and central Mexico. Under their control, war became necessary to manage the
satellite communities subjected to Paquime's authority. Corrected dates of occupation
for Paquime place its florescence after the fall of the Toltecs (Dean and Ravesloot 1993).
Hence, the Toltec pochteca explanation unravels, taking with it the warfare complex.
Subsequent publications (e.g. LeBlanc 1999; Lekson 1999a) are based on the
warfare/massacre theory, which also brings them into question.
Armed with these new dates, many general questions answered by the 1974 Casas
Grandes site report need to be revisited, for instance: 1) why were they engaged in
warfare?, and if a massacre occurred, why?, and 2) why did the invaders flee such a
fertile valley afterward? Specific questions regarding the burials must also be addressed:
1) why are there so few burials at Casas Grandes?, 2) why is there a disparity in
representation of the sexes?, and 3) does Casas Grandes share any mortuary similarities
with other sites that might explain these inconsistencies? Resolution of these questions
with physical evidence will paint a truer picture of the occupation and abandonment of
Paquime.
8Hypotheses
Three hypotheses will be tested in this dissertation. The first hypothesis:
Violence was endemic in the Casas Grandes region. To evaluate this hypothesis, the
trauma frequency of the formally interred group will be compared to other sites. These
comparative assemblages will be ones known to have a history of warfare and those
without. This hypothesis will be confirmed if the trauma frequency at Casas Grandes is
similar to those sites where warfare has been validated with both osteological and
artifactual evidence (i.e. defensive location and structures, warrior imagery, armanlents).
The second hypothesis: The final population ofPaquime was massacred. This
hypothesis will be evaluated by comparing trauma frequencies of the formally and
informally buried groups. If I find that trauma frequencies of the informally buried group
are significantly higher than that of the formally buried group, the hypothesis will be
confirmed. If the informally buried group represents a massacre, then injury should show
no predilection for sex or age (Owsley et al.1977; Willey 1990). If all the non-interred
individuals died violently, then most or all of the skeletons will show unhealed skeletal
injuries. Fluoride ion concentration analysis and accelerator mass spectrometry
radiocarbon (AMS) dates from both groups will be compared. A terminal massacre
would be characterized by similarities in low fluoride ion concentration among the
informally buried individuals as compared to the formally buried, which should show a
pattern of continuous fluoride ion concentrations throughout the time period of
occupation. AMS dates from selected individuals will be obtained for absolute dating of
the massacre.
9The third hypothesis: Human sacrifice and cannibalism were practiced at
Paquime. This hypothesis will be evaluated by comparing the types and patterns of
trauma found at Paquime with those from sites where warfare, massacre, witch killing,
cannibalism, and simple mortuary corpse processing have been verified. The hypothesis
will be confirmed if the types and patterns of trauma at Paquime are similar to sites where
evidence of human sacrifice and/or cannibalism has been found.
Significance
In addition to DiPeso and colleagues' (1974) original site report, osteological
research at Paquime has been limited. This dissertation will add to the small body of
work by scholars interested in the biology of the people ofPaquime. It will make three
major contributions. First, it will clarify the abandonment of the city and build a better
understanding of the interactions of its people with others in the region. These
interactions may have had wider reaching consequences for groups in central Mexico and
the United States Southwest. Second, it may provide insight into the nature and patterns
of prehistoric violence in the New World and a better understanding of the prehistoric
foundations of modern violence. Finally, this research will introduce the first AMS dates
taken directly from human bones at Paquime. Though these dates are few, they will
certainly assist in more precisely revising the chronology at Paquime.
Organization ofthe Dissertation
Chapter II describes the study area and the difficulties ofdefining it
geographically. It describes what excavation has revealed about the prehistory of the
Casas Grandes region, and insight into cultural connections to the south and north.
10
Chapter III describes the burials and summarizes the application of osteological analysis
to archaeology. Brief treatment is given to prior osteological studies done at Casas
Grandes, followed by a discussion ofthe properties of bone and bone fracture, and the
challenges associated with determining cause and manner of death. Chapter IV describes
the sample selection process and methodologies employed in osteological data collection,
demographic analysis, data analysis, and chronometric dating. Chapter V presents the
results of the osteological and statistical analyses, and chronometric dating along with
results of the skeletal element representation analysis. Chapter VI discusses the results
within the context ofthe three hypotheses. Chapter VII discusses site-wide applications
of the results and the influence culture may have played in the observed burial attributes.
Chapter VIII summarizes and draws conclusions about the research and offers ideas on
future research directions related to this project. A final series of appendices presents
supplementary information, which is followed by the bibliography.
11
CHAPTER II
THE CASAS GRANDES REGION
Defining the Geographic Study Area
The term Southwest has been used to describe an expanse of north and central
America that includes parts of the United States and Mexico. Cordell (1997) describes its
north-south geographic boundaries as beginning at Durango, Colorado and terminating at
Durango, Mexico and its east-west boundaries as beginning in Las Vegas, New Mexico
and extending to Las Vegas, Nevada. The Southwest culture area distinguishes itself
from neighboring cultures by the practices of digging stick agriculture to cultivate the
staples of com, beans, and squash, the use of manos and metates (grinding stones), the
manufacture of high-grade pottery, the construction of multiroom pueblo villages and
rancherias (dispersed settlements), and the occasional development of more complex
towns with unique forms of public architecture. This culture area lacks state-level
government and accompanying social stratification, a developed writing system, and
large urban centers with monumental architecture on the scale of the Oaxacan pyramids
(Cordell 1997:4).
The geographic area often referred to as Northwest Mexico is problematic in its
inclusion in geographically inclusive terms such as "the Southwest" or "Greater
Southwest." These terms derive from their geographic position within North America.
The difficulty of affiliating Casas Grandes and other northern Mexican pueblo-type sites
with a larger geographic area lies in their enigmatic position below the United States-
12
Mexico border, and the numerous cultural similarities they share with the Southwest.
The terms "Southwest," "American Southwest," and "Greater Southwest" are
inappropriate in that they imply the exclusion of northern Mexico. Likewise, DiPeso's
term "Gran Chichimeca," which he devised to include the expanse of land on both sides
of the international border thought to have been inhabited by the Chichimec people, is a
derogatory term derived from the Nahuatl term for "lineage of the dog" (Coe and Koontz
2002; DiPeso 1974). Thus, it is culturally insensitive and will not be used.
If one looks at the cultural similarities between the pueblo sites in northern
Mexico and the United States Southwest, the international distinction would become less
obvious. A recent edited volume (Webster et al. 2008) takes this issue to task, noting the
inclusion of such terms as "North American Southwest" and "the western U.S.-Mexico
borderlands" (McBrinn and Webster 2008:3), and; "Arid America" and "Oasis America,"
(Mendiola Galvan 2008:294). However, no nomenclatural consensus is reached in the
volume.
Riley (2005) prefers the term Aztlan. This Aztec word referred to the mythical
"Place of the Herons," from which the Aztecs believed they originated. This place was
far to the north and west of the Aztec kingdom. Spanish accounts cite that the Aztecs
allegedly undertook a pilgrimage to Aztlan sometime during the reign of Moctezuma I
(A.D. 1440-1469) (Riley 2005:5). Riley (2005:6) defines Aztlan geographically as:
"generally bounded by the southern Great Plains and the northern Mexican
plateaus to the east, and the southern flank of the Rocky Mountains to the
north. Its western boundary is the lower Colorado River and Gulf of
California. To the south a fuzzy line of demarcation runs from the major
Sonoran river basins eastward across the Sierra Madre Occidental through
the high plateaus and interior basins of Chihuahua and Coahuila. In
13
modem tenns the region includes the U.S. states of Arizona and New
Mexico, as well as parts of Utah, Nevada, Colorado, California and Texas.
In Mexico it encompasses much of Sonora and Chihuahua, with influences
that reach into portions of Sinaloa, Durango, and Baja California Norte."
Further compounding the problem, Casas Grandes shares a number of similarities with
cultures of central Mexico and the pueblo cultures in Arizona and New Mexico. In this
dissertation I will consider Casas Grandes as part of Northwest Mexico, interacting with
cultures to the north and south. I prefer to think of it as part of the "borderlands" or on
the "frontier" due to its intennediate location in the expanse of land north of the great
cultures of central Mexico and south of the pueblo cultures of the Southwest. For me,
this obscures the modem international border and forces one to look at the Casas Grandes
culture with respect to its state-level neighbors to the south and the pueblo cultures to the
north. This concept seems useful as the Casas Grandes culture shared attributes of both
and should be considered unique in this respect.
The Site and Its History
The cultural phenomenon known as Casas Grandes existed from about A.D. 800
and peaked from about A.D. 1200 to about 1450 with the florescence ofthe city of
Paquime. The site is located on the vast, fertile floodplain of the Casas Grandes River in
northern Chihuahua, Mexico, approximately 120 km south of the Mexico-New Mexico
border in the relatively high Basin and Range country (Map 1). Several small rivers-the
Cannen, the Santa Maria and the Casas Grandes-are located in the vicinity of Casas
Grandes and likely attracted people to the area prehistorically. The Casas Grandes River
is the largest in northwest Chihuahua, and the floodplain within a 5 km area of the site is
14
150,
1s
N
l'
u·s
• ..q
miles
kilometers
* Archaeological S~e
o
I
o
EIPasoU. S. A.
~---.-----p,a!omas Cd. Juarez
~ -1'- __ I
\
\
I • Janos
\ Nuevo Casas
Casashrandes • Grandes
Paq1me *.
I
t
J
I
( I
I Mader:,
\
\ /
( • CHIHUAHUA /
\ /(--/. / <
\ Cuauhtemoc , / -J
\ / :J
\lI" \ ( :z::
<s>,.. "- \ "t'
,..<? l \ ~
~ r. \fo-,-~-.,,- \
/ 0 \ :....r-- ~ _ _ \
./ C'C',\ ) ,_ ""
./ " , \ I
,/ 0'<S>1 I -..../"-- '
? (- 'J "~- \
SINALOA ~ (
" J DURANGO)-_/
Map 1. Location of Casas Grandes (Paquime).
estimated to have sustained 3,000 people (Whalen and Minnis 2003:319). A nearby
natural spring and an underground well at the site provided ample potable water.
Whether the people of Casas Grandes were hunting, gathering, or farming, this land at the
eastern foot of the Sierra Madre Occidental was a very good area to forge an existence.
In fact, it may have been one ofthe best places in northern Mexico and the United States
Southwest to make a living (Whalen and Minnis 2001a:320). The modem-day people in
this area continue to exploit it for the same reasons.
15
The first record ofPaquime made by a European was that of Baltasar de Obregon,
a Spanish servant to King Phillip II, who in 1565 passed through the Casas Grandes
valley on an expedition into northern Mexico (Hammond and Rey 1928). Obregon, so
impressed with the enormity and skillful construction of the city, wrote it "seemed to
have been built by the ancient Romans" (Hammond and Rey 1928:205). He further
describes the six- to seven-story walls as being solidly built, whitewashed and painted in
color with pictures. Architectural features such as floors and patios paved with stones
resembling jasper, and the large wooden pillars supported by circular stones were also
noted. He describes the Casas Grandes valley as "fertile and beautifuL .. surrounded by
splendid and rich mountains and small mountain ridges ...the most useful and beneficial
of all the rivers we found in those provinces," (Hammond and Rey 1928:206). Obregon
talked (using signs, not words) with nearby native peoples to find out where the former
occupants had gone. He writes:
"they were settled and living six days down the river toward the north, and
that they had been forced to move away on account of the war waged on
them by their enemies who came from the other side of the mountains.
They added that about four days to the west lived many other people who
occupied houses of great height, who wore clothes, and who possessed
large amounts of cotton clothing, com, beans, calabashes, fowls, and
native cattle" [i.e. they had knowledge of Southwestern people]
(Hanunond and Rey 1928:208).
The site was excavated by the Joint Casas Grandes Expedition (lCGE) from 1958
to 1961, a collaborative effort between the Amerind Foundation of Dragoon, Arizona,
and the Instituto Nacional de Antropologia e Historia (INAH) of Mexico (DiPeso 1974).
The project was under the direction of Charles DiPeso, and in his estimation excavated
16
approximately 42 percent of the total settlement (DiPeso 1974; Woosley and Olinger
1993). In addition to the architectural features of the massive city, the excavation
unearthed abundant ceramics, nearly four million pieces of shell, the remains of turkeys,
macaws, and 576 human burials. One hundred twenty-nine of these individuals were
recovered from the surfaces of residential floors and plazas, and in room fill above the
floors rather than the customary formal sub-floor burial context (DiPeso et al.
1974:8:325). DiPeso interpreted 126 of these non-interred bodies as the mortal remains
of "women, children, old people, and men of fighting age who died at the hands of an
enemy people on the fatal day of the city's destruction" (DiPeso, 1974:2:639).
Excavation found it to be one of the largest prehistoric or protohistoric pueblos,
boasting I-shaped Mesoamerican style ballcourts, flat topped mound structures, and
multistoried apartment structures (Whalen and Minnis 2001a:3) (Figure 4). It was
composed of multi-story poured adobe apartment complexes that served as the core, with
public and sacred ceremonial spaces such as flat-top mounds, plazas, agave roasting pit~,
and ball courts built around it (Cordell 1997; Whalen and Minnis 2001a). According to
DiPeso (1974), the total number of rooms was close to 2,000, which he believed would
have housed a minimum of approximately 2,240 people and a maximum of4,700 people.
17
/
Cisf8rn
1-23 U...il designotiol'i~
@ W;Jter oreos
= Wn~er !>y!>lem
~ WoJls or unll boundaries
C3,0
South Plaza /'
0;l\l/
.f~~:!5
. Bollcourt/ .
. ~r.~OlE...!!.~ _~/!!!J!r.e.~.-/ ,"
T
Meters
r====~--'l
o 50 100
.".
."':-::. ---
'- ..
."' .
...-....~
.. ----
8
Figure 4. Casas Grandes site plan (from Dean and Ravesloot 1993).
Unit 1: House of the Ovens
Unit 2: Mound of the Cross
Unit 3: Ball Court 1
Unit 4: The Mound of the Offerings
Unit 5: Mound 1-5
Unit 6: Buena Fe phase ranch-style compound
Unit 7: Retaining wall and room
Unit 8: The House of the Well
Unit 9: Mound of the Heroes
Unit 10: Mound of the Bird
Unit 11: The House of the Serpent
Unit 12: The House of the Macaws
Unit 13: The House of the Dead
Unit 14: The House of the Pillars
Unit 15: House cluster
Unit 16: The House of the Skulls
Unit 17: Ball Court II
Unit 18: South Plaza rooms
Unit 19: House cluster
Unit 20: House cluster
Unit 21: North house
Unit 22: House cluster
Unit 23: House cluster
18
Site Chronology
The site's chronology is divided into two periods, the Viejo (A.D. 800-1200) and
the Medio (A.D. 1200-1450) (Figure 5). DiPeso's dates of occupation, which were based
on non-cutting tree-ring dates, further divided each period into phases. The Viejo period
contained the Convento, Pilon and Perros Bravos phases; the Medio period contained the
Buena Fe, Paquime, Diablo and Tardio phases. The Buena Fe phase (A.D. 1060-1205)
was characterized by the beginning of intense construction andurban renewal which
came to its pinnacle during the Paquime phase (A.D. 1206-1260), before falling into
disrepair during the Diablo phase (A.D. 1261-1340). Schaafsma and Riley (1999:Table 1)
assert there is inadequate evidence to make phase distinctions in the Medio period.
The Casas Grandes chronology has been a subject of contention among Southwest
archaeologists since it was initially proposed by DiPeso. This is due mostly to
inconsistencies between the dendrochronology and the ceramic sequence. Ironically, a
major focus of the lCGE was to build a chronology of the Casas Grandes region to aid in
placing it in the Southwest scheme (DiPeso et al. 1974:4:9). Three hundred eighty-six
tree-ring samples were collected, which produced 53 dates that span the "interval from
A.D. 1044 to 1338 (Dean and Ravesloot 1993:89; DiPeso et al. 1974:4:13). These
samples came from twenty-nine rooms in five architectural units. However, the samples
did not represent actual cutting dates as a number of the outer rings were removed when
shaping the beams and posts during construction. DiPeso made the crucial error of
assuming few rings had been removed and treated the samples as near cutting dates
(Dean and Ravesloot 1993:89). He used these assumed dates, ceramic distributions,
19
obsidian hydration dates, and radiocarbon dates from other area sites to develop the
controversial Casas Grandes chronology (Dean and Ravesloot 1993:89-91; DiPeso et al.
1974:4:13).
In 1993, Jeffrey Dean and John Ravesloot re-evaluated the Casas Grandes tree
ring dates. Using a regression equation to estimate the number of outer rings lost during
construction, they were able to estimate dates from 45 samples. These dates spanned
from A.D. 1160 to 1419. The bulk of them (thirty-nine) fell in the thirteenth and
fourteenth centuries, demonstrating that construction was concentrated during this time.
Furthermore, they believed the site may have been inhabited as late as the 1470's (Figure
5). This new chronology correlates better with the Southwest ceramic and cultural
sequences and is believed to be more accurate than the one DiPeso proposed.
Cultural Connections
Several scholars see similarities between Paquime and Mesoamerica, and the
group of sites at Chaco Canyon, New Mexico (DiPeso 1974; LeBlanc 1986; Lekson
1999a; 1999b; Lister 1978). DiPeso (1974) attributed the origin ofPaquime to the
Toltecs of the Valley of Mexico due to their contemporaneity (according to his original
dating scheme), and the presence at Paquime of several Mesoamerican elements. These
included stone disc post foundations, t-shaped doorways, colonnaded structures,
decorated effigy vessels, copper bells, macaws, ballcourts, plazas, platform mounds, and
images on pottery ofthe plumed serpent god Quetzalcoatl. Several of these elements are
found at other sites throughout the United States Southwest (Cordell 1997; LeBlanc 1986;
Whalen and Minnis 1996). These elements, associated with the Mesoamerican Mixteca-
Figure 5. Relationship of Casas Grandes with other chronologies.
20
1400 Cliff Phase
Medio Tardio Pueblo IV
1300 .9 Diablo'0
<I.)
::g Black Mountain
1200 Post-Classic
Paquime Phase
Buena Fe Pueblo III
Mimbres Phase
1000 Perms
Viejo Bravos
Pi16n Pueblo II
900
0
(Terminal) '0';> Late Pithouse
Classic Period
Convento Pueblo I800
700 Early Pithouse
A.D. (Phases) Basketmaker II Period
Mesoamerican New Casas DiPeso's Southwest Mimbres
Chronology Grandes Original Pecos Chronology
Chronology" Chronology Chronology
aafter Dean and Ravesloot, 1993.
Puebla style probably originated in the Valley of Mexico diffusing up the western
coast and into the American Southwest and Paquime via trade (Coe 1994).
The influence of Casas Grandes in the Southwest is believed to have
superceded that of the Mimbres Valley Classic Animas Phase (A.D. 1100 to 1350).
This is reflected in the ceramic assemblage containing Playas Red Incised, El Paso
Polychrome and Ramos Polychrome, which is used as evidence of a close cultural
association between the Mimbres Valley and Casas Grandes (Cordell 1984). The
shift from cobble-walled structures to adobe and the move to the south of the
Mimbres Valley may also be evidence of this (Cordell 1984).
21
It is generally agreed that Casas Grandes served as the hub of a regional
system based on trade and a common system of religious beliefs (Cordell 1984;
DiPeso et al. 1974; Douglas 1995; Whalen and Minnis 2001a). The Casas Grandes
regional system bears similarities to the Chaco Canyon system due to its central core
area and satellite communities. Likewise, both systems were linked through a system
of roads, paths, and possibly a signaling system (Corde111984; DiPeso 1974; Lekson
1999a; Swanson 2003). A minor difference is that the Chaco core area consisted of
several settlements within Chaco Canyon whereas the Casas Grandes core consists of
a single, large settlement with several smaller outlying communities (Cordell 1984;
DiPeso et al. 1974; Lekson 1999a). In terms of size, Paquime is about six times
larger than the next largest regional settlement (Whalen and Minnis 2001a).
Casas Grandes has been recognized as the primary center of the region since
the publication of the site report (DiPeso 1974; DiPeso et al. 1974). However, the
degree and extent of its influence have been the subject of recent discussion. Whalen
and Minnis' (2001a) book on the Casas Grandes region is the culmination of
extensive surveys.
They see the Casas Grandes region as operating through a core-periphery
relationship. The core-periphery model is based on Wallerstein's (1974) 'world
system' theory, which proposed that a dominant civilization (the core) subjected less
powerful groups (the periphery) to its authority. DiPeso (1974) referred to these
elements as the "donor" and "recipient" cultures. Wallerstein's (1974) ideas were
more specific to European colonialism and have been modified into the core-
22
periphery model in order to make it more applicable to societies throughout human
history and prehistory (Whalen and Minnis 200la).
Whalen and Minnis (200la:38) believe Casas Grandes was the center of the
area's economic, ritual, and political structure that manipulated the flow of material,
energy, and even people on at least a regional level. According to them, the center
emerged as the result of peer-polity exchange. The peer-polity approach uses conflict
between social groups of approximately comparable size and social complexity as the
stimulus for change and integration of diverse cultural trappings within a geographic
area (Whalen and Minnis 2003). They believe the Medio period settlements were
arranged into three levels of interaction: the Inner, Middle, and Outer Zones. The
Inner Zone extended about 30 kIn around Casas Grandes and was the area ofmost
intensive interaction, characterized by the occurrence of the highest concentration of
large to very large communities, the most ritual architecture, and the most exotica.
The Middle Zone extended to the northwest of Casas Grandes to a distance of 30-60
kIn and is characterized by smaller communities, a lower concentration of large
settlements and the absence of very large settlements. The Outer Zone extends 100-
150 kIn farther north and west to the Carreras Basin, just below the southwest comer
of New Mexico (Whalen and Minnis 200lb:3l8-3l9).
Though viewed by DiPeso (1974) as a brutal centralized authority over all of
northern Mexico and the American Southwest (an opinion rooted in: his belief that
Paquime operated as an extension of the Mexican Toltec state), Whalen and Minnis
(200la, 2001 b) have proposed an alternative view based on their region-wide survey
23
data. Their rank-size distribution comparison of sites in the Inner and Middle Zones
shows consistencies with different levels of integration. Rank-size graphing is more
thoroughly discussed in Whalen and Minnis (2001 a). This comparison showed that
the Inner Zone settlements, including Paquime, bore a distribution characteristic ofa
settlement system made up of a regionally dominant center with a relatively
independent, loosely organized adjacent area and settlements that lacked significant
integration. The Middle Zone's distribution was more suggestive of a low degree of
organization and integration lacking a centralized settlement.
Sociopolitically, Paquime is characterized as a mid-level society (Whalen and
Minnis 2001a), which shares similarities with ranked chiefdoms (Bailey and Peoples
1999). Mid-level societies are characterized as having developed beyond
egalitarianism, but lack formal stratification and rigid decision-making hierarchies
and bureaucratic authority (Whalen and Minnis 2001a). While it appears there was
no rigid social stratification, there were likely people of higher status who regulated
the flow of prestige and trade goods, and who were in charge of urban planning and
organizing labor for such public projects as the building and maintenance of the water
distribution system and ceremonial areas of the city. These differences in status are
reflected in the uneven distribution and quality of burial goods (Ravesloot 1988).
Paquime's Genesis and Decadence
Much of the confusion surrounding the origin and affinities of Paquime lies in
the presence of a mixture of cultural elements from both the United States Southwest
and Mesoamerica. An early account by Bandelier (1890) attributes the origin of
24
Paquime to the Southwest, an idea that pervaded the archaeological literature of the
early 20th century (Brand 1935; Carey 1931; Ekholm 1940; Kidder 1916; Lister 1958).
Brand (1935:287) went so far as to say "the prehistoric or archaeologic cultures of
Northwest Mexico are definitely Southwestern in affiliation," due to their lack of
Mesoamerican cultural elements as identified by Kroeber (1928). It is difficult to
escape the fact that pueblo style adobe architecture characterizes the city ofPaquime
and its contemporaries in northwest Mexico (Cordell 1997; Whalen and Minnis 2003).
Mesoamerican Influence
It is widely recognized that Paquime was a ceremonial center that produced
and facilitated trade of exotica, much of it associated with religious rites. Much of
DiPeso's argument for ritual human sacrifice rests on the discovery of ceremonial
ballcourts and recovery of ritual paraphernalia at the site. The ballcourts may figure
prominently into arguments that human sacrifice and cannibalism were operating at
Paquime.
The ballgame was important to the ancient Mesoamericans for many reasons.
It was used as a stage for performances of rituals and sacrifice, and as a means for the
acquisition and consumption of goods (Fash and Fash 2007:267). Players acted out
dramas tied to fertility and the agricultural cycle, and for settling disputes between
communities (Fash and Fash 2007:270-271). A contest that marked a sacred event
served to stimulate production, acquisition, and consumption of sacred and secular
goods in the same fashion as sporting events do today (Fash and Fash 2007:278). It
was truly a multipurpose event.
25
Certainly Paquime was a center for the collection and production of exotic
materials, though the degree to which the city exported these items is still contentious.
The abundance of west Mexican goods shows the city was part of the extensive
Aztathin trading system (DiPeso 1974:2:625-629; Kelley 2000; VanPool and
VanPoo12007: 135), a system that could have served as the vehicle for diffusion of
the ballgame up the west coast of Mexico and east to Paquime (Wilcox 1991:105).
The VanPool's (2007) argue that an emerging ruling class developed a complex
ritual-religious system to reduce the increased scalar stress that resulted from the
growing settlement and the demands on the population as the users and distributors of
the region's exotica (macaws, shells, and copper bells, minerals). This ruling class
synthesized the existing religious structure with imported symbolism, creating a
system that utilized exotica, the Mesoamerican ballgame, tobacco shamanism, and
esoteric knowledge of how the universe operated (VanPool and VanPoo12007:133).
The elites centered themselves as performers of rituals and intermediaries between
the natural and supernatural worlds, images of which are depicted in abundance on
Casas Grandes polychrome pottery (Davis-Salazar 2007:197; VanPool and VanPool
2007).
Davis~Salazar (2007: 198) argues that accumulation and control of ritual
objects in the hands of a few elites could legitimize their power and further
distinguish them socially. As performers of rituals, the elites or shamans (these could
be one in the same) separate themselves from the audience through their control of
knowledge of sacred rites and control of ritual objects (Davis-Salazar 2007: 199). The
26
implementation of the ballgame would have been an ideal way to integrate the region
and provide a stage on which to perform rituals or dramas in front of a regional crowd
(Wilcox 1991). Thus, the ballgame would have been important to the elite of
Paquime because: 1) it provided a means for dispute resolution and social integration
between regional groups, 2) it served to bring people and their goods together for
trade and/or redistribution, and 3) it legitimized their power and reiterated the
distinction between the elites and the lesser social classes.
The ballgame may have served an integrative ceremonial function in the
region (DiPeso 1974:2:414; Riley 2005:131). Whalen and Minnis (1996) recorded 12
additional ballcourts in their survey of the valleys and foothills of the Sierra Madre to
the west. Seven more have been recorded at more distant sites-four in northeast
Sonora and three in the Animas region of extreme southwest New Mexico (Riley
2005:129-131). DiPeso (1974:2:415-415) advanced the theory that, like further south
in Mesoamerica, these games were closely tied to human sacrifice. In the center of
the T-shaped ballcourt at Paquime, DiPeso found a multiple burial below the court's
surface that contained the skeletal remains of two males buried in a position he
believes was consistent with a Veracruz palma design depicting fertility symbolism.
Beneath the surface of the south and north ends of the same ballcourt, he recovered
burials containing two females (one of them pregnant), and two females and a male
skull, respectively. At least one of the females may have suffered some post-mortem
ritual dismemberment of her right arm, which was found draped over her head.
27
Development and Emergence ofPaquime
The emergence of the city remains poorly understood. A number of theories
about its rapid development have been proposed, among them: it was founded and
developed as an outpost by the Toltecs (DiPeso 1974), or Aztec merchants (Riley
2005) to facilitate the flow of luxury items (turquoise, copper) between the southwest
United States and the Valley of Mexico; it was founded by southward-migrating
Mogollon peoples who were fleeing drought and sociopolitical upheaval (Lekson
1999a), and; it developed in situ bolstering the positions of local leaders, while
borrowing cultural elements from outside the area (Whalen and Minnis 2003).
DiPeso (1974:2:299) attributes the rise ofthe Medio period city of Paquime to
economic expansion of the Toltec Empire from the Valley of Mexico. He believes
this expansion was facilitated by an emerging class of Mesoamerican
merchants/paramilitary spies known by the Nahuatl tenn pochteca (singular
pochtecatl) (Riley 1993:17). He (1974:2:290) believed that a few Toltec pochteca
came to the Casas Grandes valley around A.D. 1060 to exploit its rich mineral and
metals resources. In doing so, they inspired the locals to build the city of Paquime
and established it as a great frontier trade center, sending such goods as turquoise and
copper to Mesoamerica and sacred turkeys and macaws to the Southwest.
DiPeso supports his model with three secondary, headless human burials from
the Mound of the Offerings, which he believes were those ofpochteca (DiPeso et al.
1974:8:335). DiPeso and colleagues (1974:8:335) recount Spanish explorer
SahagUn's description of a pochtecatl death ceremony from Aztec times, where the
28
deceased is dressed and adorned with body paint, then carried to a mountain top
where his body was consumed by animals. Afterwards, the remains were carried
back and buried. McGuire (1980) and Ravesloot (1984) question this conclusion at
Paquime, noting that the artifacts found with these burials were local in origin rather
than Mesoamerican as would be expected. Moreover, the mortuary treatment of these
high status individuals must be examined in terms of the Casas Grandes total
mortuary program rather than only by the quantity and quality of grave goods
(Ravesloot 1984:212).
In Riley's (2005: 117) view, "Paquime represented the most overt
Mesoamericanization in all of Aztlan." He believes Paquime shows continuity with
Mesoamerica in religious and sociopolitical systems, sophistication in construction
techniques, mound structures, colonnaded structures, a water system (called an
acequia) and a sewer system. Probably the most important feature was the presence
of Mesoamerican-style ballcourts found throughout the region, three of which were
found at Paquime (Riley 2005: 131).
Lekson (l999a) sees an ancestor-descendant relationship between the Chaco
communities and Paquime. He believes the Casas Grandes valley was
underpopulated relative to its carrying capacity during the Viejo period. Thus, the
seemingly instantaneous rise in population is attributable to a sudden influx ofpeople
moving southward from New Mexico. This is given validity by the extensive
movements of Puebloan peoples during the 13th century due to increasingly dry
climatic conditions (Dean 1996; discussed below). Lekson believes the original
29
inhabitants of Chaco Canyon moved directly north to the site of Aztec due to climate
change. Social and political factors forced them to leave Aztec after a short period of
time. They then moved directly south of Aztec and Chaco to the Casas Grandes
valley along a sacred north-south meridian that would forever bind them to their
homeland.
Jeffrey Dean's work on tree-ring analysis offers convincing evidence for
Lekson's model. Dean's (1996) reconstruction of long-term paleoenvironments in
the Southwest using principal component analysis of tree-rings provides crucial
information about annual precipitation and stream flow in the northern Southwest.
Dean found unimodal and bimodal precipitation patterns that repeated over the period
A.D. 966 to 1988 with the exception of the period A.D. 1250 to 1450, "when the
long-term pattern broke down into chaotic distributions ofthree or four principal
components that exhibit no logical geographic patterning" (Dean 1996:43). The
northwest component in Dean's study (the northwest area of the Four Comers)
experienced the greatest disruption of this pattern and sites in this area were
abandoned between A.D. 1250 and 1450. This time period and geographic location
correspond to an increase in violence and cannibalism (Turner and Turner 1999). The
southeast component was little changed and the sites in this area continued to be
occupied through this time. Therefore, it is possible that sites to the south such as
Casas Grandes, which were outside this chaotic rainfall pattern, saw tremendous
population growth during this time period due to the southern migration of Mogollon
peoples (Cordell 1997).
30
Whalen and Minnis (2003) believe Casas Grandes developed locally, pointing
to the availability of copious arable land in the Casas Grandes River valley, an
abundance of local floral and faunal food resources and a previously undocumented
continuity of occupation in the area. Whalen and Minnis (2003:327) call it "the most
productive agricultural setting in the region." The Tinaja site 20 km southwest of
Casas Grandes, which dates to the early Medio, shows early development of many
architectural elements previously thought to be unique to Paquime. Thirteenth
century components of the Tinaja site such as the I-shaped ball court, T-shaped
doorways, stairs, and poured adobe house construction show striking similarities to
those at Paquime. The site demonstrates continuity in building style and fills the
thirteenth century temporal gap noted by Lekson (1999a) during which it was thought
there was an occupational hiatus.
To summarize, Casas Grandes existed as a cultural phenomenon from the
thirteenth to the fifteenth century. The Medio period city of Paquime incorporated
architecture and technologies characteristic of both the Southwest and Mesoamerica.
Several viewpoints have been offered to explain the florescence of the city of
Paquime. While no consensus has been reached, the local development model is
strongly supported by the extensive survey and excavation work by Whalen and
Minnis (2001; 2003), while a mass exodus from the Southwest (Lekson 1999a) is
supported by climatological work by Dean (1996). However, Lekson's (l999a)
sacred meridian theory is difficult to support archaeologically. DiPeso's (1974)
Toltec pochteca model has fallen out of favor due to improved dating and the insight
that Raves1oot's (1984) analysis has offered. Paquime does show material culture
likely derived from central Mexico, but whether it was the Aztlan of the Aztecs, as
Riley (2005) claims, is still speculative.
31
32
CHAPTER III
REVIEW OF OSTEOLOGICAL LITERATURE
The Burials
The excavation recovered 652 individuals, 576 of which were identified as
dating to the Medio period (DiPeso et al. 1974:8:360). Period determination was
made by their association among the building structures and ceramic assemblages.
DiPeso recorded twelve main burial types at Paquime, many of which were further
divided into sub-types (Figure 6). DiPeso's burial coding method used a bipartite
system. The first number was the individual burial number from a unit; the second
number was the unit designation. For example, burial 27-14 is the 2ih burial
recovered from unit 14. Humans were typically wrapped in blankets or sleeping mats
and buried under the floors of occupied houses or plazas. Differences in social status
were reflected in grave accompaniments, which ranged from ceramics and jewelry to
tools and macaw and turkey carcasses (DiPeso 1974:2:650-651; Ravesloot 1984;
1988). Nearly one half ofthe room burials and one fourth ofthe plaza burials
contained ceramic vessels. Some vessels contained cultivated or uncultivated food
remains. Burial practices consisted of a mixture of single and multiple, primary and
secondary burials. A primary burial is distinguished as beginning and ending with the
initial disposal ofthe body (Schroeder 2001 :82). This is the typical modem burial
method in North America. During secondary burials, the human remains are removed
33
from the initial resting place and replaced either in the initial disposal facility or
removed to a second disposal site (Schroeder 2001 :82).
Taphonomic indicators of secondary burial are the absence of small hand and
foot bones, under-representation of normally well-preserved elements, fragmentation
and lack of complete elements, and smaller than expected numbers of teeth in skulls
(Roksandic 2002: 109-110). Evidence of cutmarks for the purpose of soft tissue
removal, dry fractures and canid gnawing may also be present (Redfern 2008:293).
The Medio period burials showed no apparent positional or postural preference, with
an unequal mixture of cardinal orientations, and of flexed, prone, and prostrate
positions. Five individuals were buried secondarily in three urns.
A limited number of studies have included the Casas Grandes skeletons in
their data sets. The only studies of the Casas Grandes osseous remains are found in
doctoral dissertations by Benfer (1968), Butler (1971), and Walker (2006), and in
publications by Weaver (1981) and Woodall (1968). Ravesloot's (1984, 1988) study
ofburial practices and social status is also noteworthy.
Benfer (1968) provided the first analysis of the Casas Grandes skeletal
material. His goals were to characterize the demographics of the Viejo and Medio
period individuals and determine their biological affinity with others from Mexico
and the Southwest. Unfortunately, he only included individuals between the ages of
20 and 50, due to the inability of aging techniques to produce solid age
34
ROOi\1 BU1?IALS 278 BODIES 157 PITS ~ PLAZA BUR1ALS. 135 UQD1ES 9S PLTS
TYPE I ROOM SUBFLOOR. SEALED
~.:,..:::=. ... :
.. - 0 .:- ..
....... • ... D
TYPE XI ROO~ SUB-
FLOOR.~SBALED
TYP·E IV ROOM SUB-
FLOOR,TOMB
:"~," -.. 'I.• .. f.";)
~ ." .. .. <10 •
TYPE m PLAZA SUB- TYPE XII PLAZA SUB-
FLOOR, SEALED FL.OOR, UNSEALED
4 BODIES
I.PIT
2 BODIES
I PIT
1 EODY
1 FIT
XlA tVA lIlA . xnA
2 BODtES 2 BODIES 73 Booms 5 BODIES
2 PITS 2. PITS 73 PITS 5 PITS
XlB IVB lIlB
22 BODIES 4 BODLES 35 BODIES
8 PITS 2 PITS 10 PITS
XlF
2 BODIES
2 PITS
4 BODIES
1 PIT
.2-BODIES
~ PITS
23 BODIES
7 PITS
"10 BOnq;':S
4 PITS
?1 BODY·
1 PIT
100 BODIES
35 PITS
87 BODIES
87 PITS
IA SINGLE ARTICULATED
m MULTIPLE ARTICULATED
Ie SINGLE" SE<;:ONDARY
IP M?LTIPLE SECONDARY
IE MULTIPLE ART[CULA1"ED & SECONDARY XTE
IF' BODY REMOVED
10 MULTIPLE ARTICULATED WITH
EXTRA A.RTICULATED PARTS
18 ARTICULATED BODY. PARTS REMOVED
MT<;( l' J I ;'\l'.Frll.S f3IIRIAI S \<"1 nrF.:s :2'-> p,r s
TYIP:e V"Jl SUFEl.'tlM_ 'T'Y"PE V':IJ:[ UUf05AN TYE"J;: V TOMB. TYPE v:l 'URN.
i:~:i;. S~::~~:.:~ ·;~(~2._..
1.3 BODUU:::
13 PITS
2. aODl~S:
2 PJ-rs
3 BODJEs
3 UR,.,:']
2. aODlJ!:-S
1 URN
XJIU\. FI"lOM
UNSEALED
P~OVENt-ENCI$
S El<DD1ES
2 PITS
3 OODLES
1 prT'
XLrrB FF't0311
S ........L..... D
'p~OV~l"IIENC~
XfilC ~P:OM
ARCH1TI::tCTUR .... f_
DEBR1S
Figure 6. Casas Grandes burial types (from DiPeso et al. 1974:Figure 358-8).
35
detenninations above 50. He provides no explanation for excluding the subadults, but
concedes he may have been remiss in doing so stating, "that project, along with others,
remains for future workers" (Benfer 1968:2). Since Benfer's focus was on population
affinity, his treatment of pathology and trauma was limited with the exception of
describing the trophy skulls.
Butler (1971) described the crania and dentitions of the Casas Grandes
assemblage across the Viejo, Medio, Tardio, and Spanish contact periods. She
hypothesized that if small groups ofpochteca from an outside "donor" culture did
indeed inspire the indigenous people to build Paquime as a Mesoamerican outpost,
then the skeletons would show heterogeneity in their expression of inherited discrete
traits. Her results showeq. no evidence for populiltio.p hetero~eneity until
~nRmf'imMe~y tR-r Hme o~SPflnish contact.
Butler included 306 of the adult Medio period skeletons (those over age 18) in
her analysis. Her totals by sex were: 104 males, 148 females and 54 of unknown sex.
Thirty-seven. of the 54 of unknown sex were also of unknown age. This differential
representation of sexes is the evidence DiPeso used to speculate a warfare cQmplex
existed, and that the missing males were lost on distant battlefields or buried in a
separate, yet unexcavated warrior cemetery.
Woodall (1968) indexed nutrition and morbidity in the Casas Grandes
skeletons by documenting the occurrence of Harris growth disruption lines in tibias.
Harris lines are thought to represent periods of disruption in nonnal growth. He
fpund seven pfthe thirty-nine individuals in the study showed the transverse lines.
36
Though his results were largely inconclusive, it did show one of the possibilities for
skeletal research.
Weaver (1981) investigated diet change between the Viejo and Medio periods.
He hypothesized the people would suffer a decrease in health that corresponded with
the transition from foraging to agriculture. By recording the frequency of occurrence
of porotic hyperostosis and periostitis, skeletal indications of nutritional deficiency
and disease, he concluded there was a significant difference in their occurrence
between the Viejo and Medio period populations. He believed this was due to a
decreased standard of living with the adoption of agriculture. However, Corruccini
(1983) points out inconsistencies in Weaver's chi-square analysis of the samples,
which when corrected showed no significant differences between the frequencies of
these pathologies. Thus, the study shows results similar to Woodall's (1968).
Walker (2006) investigated biological affinities of individuals from cave
burials he excavated from the Pima and Tarahumara regions of southwestern
Chihuahua, Mexico. While his focus was not on Paquime, he did include a sample of
the Medio period skeletons. He found some long-tenn continuity between the
individuals from the cave contexts and the modem inhabitants of the area. However,
despite the temporal overlap of these burials with those from Paquime, he found little
evidence to suggest continuity among the two.
Ravesloot (1984; 1988) studied the influence of ascribed social status on
mortuary practices at Casas Grandes. He examined the total mortuary program (body
preparation, burial facility, location of burial facility, quantity and quality of grave
37
goods) using principal component analysis rather than by the number and quality of
funerary objects alone. He found differences in the mortuary practices that showed
Casas Grandes may have been organized by ascriptive ranking, suppOliing DiPeso's
assertion of a social hierarchy. He also noted that many ofthe non-interred skeletons
were processed in a manner similar to the formally interred ones (plaza subfloor
unsealed grave pits, plaza fill, room fill, and single interment). He writes: "This may
reflect a digression in the organizational complexity of the community near the time
of its decay and abandonment. Or, it may represent another dimension of status
differentiation operating during the Medio" (Ravesloot 1984:207; 1988:72-73).
Ravesloot (1984) proposes an alternative to the pochteca political model
based on his mortuary study. He writes the Paquimeans may have been organized on
the basis of ancestor worship where
"an individual's placement within the social hierarchy at Casas
Grandes was presumably assigned at birth on the basis of his or her
genealogical relationship to these ancestors. This genealogical
relationship determined social positions accessible within the status
hierarchy including ones that carried political, economic, and religious
authority or power" (Ravesloot 1984:213).
He supports this proposal with evidence that few individuals at Casas Grandes were
given elaborate ritual treatment, which included burial in special tombs in special
locations of the city. Many of these tombs contained multiple burials of both sexes
and various ages. Several of these tombs are associated with ceramic drums and
headless turkeys, both of which are associated with formalized funerary rituals. In
the end, Ravesloot (1984:3) stresses the value of re-examining archaeological
material from alternative perspectives.
38
In summary, skeletal studies of the people of Casas Grandes show a lack of
nutritional deficiencies during the transition to agriculture, a low disease load,
introduction of foreign genes was negligible until Spanish times, and they show no
biological relationship to extant local populations. These studies demonstrate the
promise that skeletal analysis holds for increasing our knowledge of the culture and
biology of the people of Casas Grandes. However, the material still holds a wealth of
untapped information. Specifically, none of these studies address skeletal trauma, the
concern ofthis research. With the exception of Ravesloot's (1984) analysis,
exclusion of the non-interred skeletons serves as a common thread.
The Nature and Properties ofBone
The skeletal system constitutes less than 20% of the weight of the human
body but functions as one of its most important systems (White and Folkens 2005). It
protects and supports soft tissue, anchors muscles, tendons and ligaments, acts as
levers to produce movement, and provides centers for the production of blood cells
and storage of fats and minerals (Baker et al. 2005:5). Bone is composed of two
basic components known as compact or cortical bone, and spongy (also called
trabecular or cancellous) bone. Their composition is the same at the cellular and
molecular levels; the only difference is in the degree of porosity (White and Folkens
2005). Bone is a composite of collagen and hydroxyapatite, the organic and
inorganic components, respectively. The organic component provides the rigidity,
hardness, and resistance against compression (Baker et al. 2005:5). Collagen makes
up about 90% of this organic content. The bones of children are comprised of mostly
39
organic material, which gives them greater elasticity, flexibility, and resistance to
breakage than those of adults. This lack of mineralization also makes them less likely
to be preserved in archaeological contexts. Much of the organic fraction is replaced
by inorganic matrix with advancing age.
According to Berryman and Symes (1998), fresh bone is a viscoelastic
material. Due to its viscoelastic properties, bone is nearly twice as strong in
compression as in tension. It defonns under load but has the capability to restore its
original shape once the load is lifted. If the load exceeds the bone's elastic
capabilities it will become pennanently defonned even after the load is lifted. Hence,
when the bone's elastic and plastic capabilities are exceeded it will fail. Once the
bone dries after death this elastic property is reduced, causing easier breakage, which
explains why archaeological bone subjected to the compressive forces of burial is
often recovered in a fragmented state.
The Timing and Types ofBone Fracture
Before discussing damage to bone, a distinction must be made between bone
fracturing and splitting. According to Noe-Nygaard (1977), bone fracturing results
from cutting, beating, breaking, or crushing the bone, either accidentally or
purposefully. Intentional bone splitting is distinguished by the desired end result of
splitting a bone longitudinally for easier and greater access to the marrow contents
during animal processing. This is accomplished by intentionally striking the bone
diaphysis with a series of precise blows along its longitudinal axis. Noe-Nygaard was
able to demonstrate experimentally that this is best accomplished after removal of soft
40
tissue from the bone. Archaeologically, soft tissue removal would have been
accomplished by passive exposure, removal following exposure to heat (fire or
boiling), or by cutting it away with a sharp tool.
Damage to bone is classified by its timing relative to the time of death. The
three classifications are: antemortem (sometime before death), perimortem (around
the time of death), and postmortem (after death). Antemortem fractures show some
degree of bone callous formation or complete healing (Sauer 1998:322). In cases of
depressed fractures of the skull, the margins of the fracture will heal, but the
depressed area remains throughout life (Walker 1997).
Perimortem injuries can undergo some degree of healing, but may lack
complete healing as it can take approximately two weeks before healing begins, and
the affected area can remain plastic for up to two months (Roksandic et al. 2006;
Sauer 1998:325). These can be recognized archaeologically by their sharp edges,
and the presence of adhering bone fragments. Pickering and colleagues (2005) report
that breaks to fresh bone are more acutely angled (relative to the long axis ofthe bone)
than similar breaks to dry bone. This is due to the presence of a higher organic
fraction in the bone at the time of breakage and the use ofpercussive force (Pickering
et al. 2005:248).
Postmortem fractures that occur to dry bone are relatively easy to recognize as
the edges are sharp, there is no evidence of healing, the angle of fracture in longbones
is less acute, and adhering bone fragments are absent (Pickering et al. 2005:248;
Sauer 1998:325). Often, postmortem fractures are the result of recovery techniques
41
and the edges of these breaks are distinguished by a difference in color (White 1992).
I will use this classification in this dissertation, though fractures to wet bone after
death will be considered postmortem.
Cause and Manner ofDeath from the Skeleton
When confronted with human remains, researchers are often interested in
determining the cause and manner of death. Cause of death is the disease or injury
that caused the person to die (i.e. heart attack, stroke, asphyxiation), whereas manner
of death concerns the circumstances of how the death occurred (homicide, accident,
suicide, natural causes, undetermined) (Clark et al. 1996:185). Though violent death
often leaves dam'age to the skeleton making cause of death more apparent, manner of
death can be more difficult to determine and often relies on other factors related or
unrelated to the body (time-since-death, eyewitnesses, position of the body, trace
evidence, associated artifacts, for example).
Violent deaths often leave skeletal damage from which a researcher may infer
the cause of death. Depressed cranial fractures are often recorded as a type of injury
attributed to interpersonal violence that occurred in prehistory. Blunt-force
instruments such as stone tools and clubs are likely to cause these localized depressed
fractures that lack radiating cracks. Severe depressed cranial fractures can be highly
lethal, whereas less severe ones may be non-lethal if they do not penetrate the inner
table of the cranial bone and into the skull cavity. Sharp-force injuries attributable to
a spear or arrow can be differentially diagnosed because they often leave the stone
point or part of it embedded in the bone (Walker 2001:Figure 1).
42
Determining manner of death from skeletal remains from modem or
prehistoric context can be difficult if not impossible. Once skeletal damage is
identified, one must differentiate between accidental and intentional injuries. Several
studies have found that skulls of those engaged in recurrent bouts of war showed
multiple healed and in some cases unhealed depressed cranial fractures (Lambert
1997; Torres-Rouff and Costa Junqueira 2006; Tung 2007; Walker 1989). The
percent of those affected by war-related trauma was lower (24-35%) than those found
in massacre assemblages.
Differences in injury location and among age groups and sexes are also
informative. Two studies found that over half ofnon-lethal cranial injuries were to
the frontal bone, which suggests face-to-face conflict (Lambert 1997; Torres-Rouff
and Costa Junqueira 2006). Furthermore, the greater occurrence of injury on the left
frontal ofmales suggested the aggressors were predominantly right-handed. Cranial
injuries to females were more diffuse and were primarily located on the posterior
portion of the skull, which is more suggestive of spousal abuse or some other form of
violence rather than warfare. Lambert (1997; also Walker 1997) found that healed
fractures were much more common in males than in females. Healed cranial vault
injuries were most common in those aged 18 to 40, while children under ten years and
those 40 years and older were infrequently affected. This suggests that most injuries
were incurred during the adolescent and young adult years, and that those who were
not involved in the violent episodes lived longer.
43
Fractures of the distal ulna, the so-called "parry" fracture, can indicate
defensive wounds resulting from interpersonal violence. The name "parry," coined
by Elliot Smith and Wood Jones was used to describe fractures of the distal ulna that
were thought to have occurred in an individual's attempts to deflect blows from an
incoming weapon (Jurmain 1999:215). However, this cause is equivocal as some
studies have found that fractures of the distal ulna are common in non-warring groups,
and equally distributed between the sexes (judd-Roberts 1999; Jurmain 1991; Neves
et al. 1999). Isolated cases lacking other trauma are more likely the result of
accidental falls (Smith 1996:84). However, when found together with other injuries
such as those of the skull, this type of fracture may be due to interpersonal violence
(Tung 2007:945).
In any case, physical and circumstantial evidence should be treated as the
primary basis for making a decision rather than drawing conclusions based on myth
and speculation. I will illustrate this point using three brief examples. The collapse
of the ancient city of Mohenjo-daro is surrounded by myth. Dales (1979) argues that
while South Asian lore indicts Aryan invaders for the fall of Harappan civilization,
the lack of mass casualties from both sides makes it difficult to accept. While there
are some scattered human remains at Mohenjo-daro, their small numbers (37
individuals) and imprecise dates make it difficult to attribute them to one time period
much less one single event. Kennedy (1984) also takes this position after examining
the skeletal remains from Mohenjo-daro and finding conclusive evidence of only one
44
case of traumatic death. While this death was attributed to violence, the identification
of one victim is not evidence of a massacre.
Room 38 in unit 11 at Paquime serves as the second example. The
stratigraphy of Room 38 shows layers of room fill containing macaw remains,
fragmented human remains, and several stones inside the doorway. DiPeso
(1974:5:476) contends the function of this early Medio period room was sacred.
Later, it was converted to a utilitarian space, during which time a macaw aviary was
constructed on its roof. A drawing of the homed serpent god Quetzalcoatl is
attributed to a youth who doodled on the wall while tending the macaws up until the
time the roof collapsed, killing both the boy and the birds. The room was
subsequently used as a cemetery for later occupants of the city. Finally, the last
inhabitants barricaded themselves in the room during the massacre, sealing the entry
with stones. The invaders pushed through the door, sending the stones down the
ramped entry before killing the occupants.
Using the same data, Walker (2002) questions the homicidal manner ofdeath
of these individuals. Citing similar stratigraphic sequences at kivas in the Southwest,
he proposes the sequence better represents the construction of a kiva, a ritual closure
of its ramp, and its remodeling into a more specialized room before being used as a
cemetery space. The macaws (a highly sacred animal) were reared on the roof and
the drawing of Quetzalcoatl was intentionally made in the sacred room. Finally,
macaw and human remains were deposited in the room during the ritual closure.
Thus, the unrelated sequence of events proposed by DiPeso lacks parsimony when
45
declaring an accidental death and subsequent massacre occurred in this room. I think
it is also possible that this room was modified to serve as a burial crypt rather than a
kiva (Weiss-Krejci 2001 :778).
The final example illustrates how occasionally, not only are cause and manner
of death uncertain, but also the identity of the victims. Turner and Morris (1970)
analyzed skeletal remains from Polacca Wash, New Mexico, which they believe to be
the remains of captives taken after the eighteenth century massacre at Awatovi.
Relying on oral histories as the primary source of information regarding witchcraft
and violence in the Southwest, Darling (1998) believes the remains at Polacca Wash
are those of people killed following the Awatovi massacre. When rebuking Turner
and Morris' (1970) assertion of cannibalism at Polacca Wash, he cites stories that
often do not agree with each other depending on the source. According to legend,
intercommunity suspicions of witchcraft surrounded the people of Awatovi. One
night, warriors from nearby pueblos converged on Awatovi where most of the men of
the village were in kivas. The warriors removed the ladders from the entrances as the
underground structures were ignited. In one version of the story, the women and
children were taken captive and later tortured, killed, and dismembered at other
locations. In another, a group of surviving warriors from Awatovi tried to rescue the
captives, but were themselves taken captive, tortured, killed, and dismembered at a
different location.
The location of the "death mound" where the captives were killed is not
certain. Turner and Morris (1970) found that when talking with local people, several
46
mounds associated with killing events were strewn throughout the area, and their
associations depended on the point of reference of the informant. Hence, no one is
sure of which mound corresponds with the Awatovi captive massacre, and whether
the Polacca Wash remains are those of the captives.
Case Studies
How does one distinguish between skeletal damage resulting from warfare,
massacre, witch killing, mortuary processing or cannibalism? Each of these different
activities leaves signature damage types and patterns on bone, and when viewed in
concert with other archaeological data can render a tenable assessment. Owsley and
colleagues (1977) examined the skeletal assemblage from the Larson site on the Great
Plains. They excavated the village cemetery and several living areas where multiple
bodies were found buried under roof fall. They found that 41% of the skulls
recovered from the living areas showed evidence of scalping. This figure is an
estimate and may actually have been substantially different as an additional 31% of
the skulls recovered were too damaged to be evaluated. The overall rate of trauma of
the living area population was much higher than that found in the cemetery
population. They also found widespread evidence of blunt-force cranial trauma,
decapitation, dismemberment and crushing of the face. They concluded that the
individuals recovered from the living areas had died in a massacre and were
subsequently mutilated. They were left in the lodges, which were burned and
collapsed on themselves.
47
Willey's (1990) analysis of the Crow Creek, South Dakota skeletal
assemblage documents a dramatic example of skeletal evidence of a large scale
massacre followed by corpse mutilation. The 486 skeletons appear to be the remains
of occupants of a Plains village that were massacred. Over 90% of the preserved
frontal bones showed cut marks, with 42% ofthe crania showing depressed fractures,
mostly of the frontal and parietal regions. Willey noted cases of decapitation,
scalping, nose removal, and possibly tongue removal, identified by consistently
located cutmarks. Post-cranial cutmarks were rare, but chewing to the ends of bones
was abundant, suggesting that scavengers were the source ofdismemberment.
Ogilvie and Hilton (2000) studied a collection of skeletal remains recovered
from a kiva at Ram Mesa, New Mexico. They describe blunt force cranial trauma,
green fracturing oflongbones, cutmarks to the skull and mandible, avulsion ofthe
anterior teeth, and a high incidence ofbuming. Notably, cutmarks did not occur on
articular surfaces or at major muscle attachments of bones, which would indicate
dismemberment. The authors suggest the skeletons were the remains of witches, and
the damage was the result of their being murdered and their bodies rendered
umecognizable. This interpretation is based on earlier work by Darling (1998) and
the "pervasiveness of witchcraft beliefs in recent times" (Ogilvie and Hilton 2000:42).
Pueblo witches were believed to be possessed by malevolent powers and the
performers ofhuman sacrifice and cannibalism (Darling 1998:737). Punishment for
witchcraft was either through banishment or execution. In cases of execution, the
body was rendered umecognizable so its spirit could not re-enter it. Darling
48
(1998:735) cites ethnographic accounts of how spirits were released from the witch's
body by pounding it with large stones or pieces of wood. This method would
certainly have caused extensive fracturing. Though no standards for comparison have
been definitively established, Ogilvie and Hilton (2000:46) argue that witch execution
is the most parsimonious explanation for the pattern and extreme degree of corpse
processing seen at Ram Mesa coupled with the deposition of the bones in a kiva.
Nelson and colleagues (1992) analyzed the remains of 14 individuals
deposited on the floor surface of a room at La Quemada, Zacatecas. The bones were
found in three concentrations within the room and the skeletons appeared not to have
been articulated at the time of deposition. They found no cutmarks that would
indicate dismemberment or mutilation, but there was a lack of vertebrae and bones of
the hands and feet. They concluded the bones were those of individuals who died,
their bodies left to decay naturally and the bones gathered and deposited in the room,
which likely functioned as a charnel house. Under-representation of small and
cancellous bones is likely due to diagenesis and poor recovery of small bones by the
persons who deposited the bones in the room.
Following the article by Turner and Morris (1970) several investigations have
documented cannibalism at sites primarily in the Four Corners area of the Southwest
(for a summary see Turner and Turner 1999). Similar evidence has been recorded at
Old World archaeological sites in Spain (Caceres et al. 2007), Algeria (Haverkort and
Lubell 1999) and France (Villa et al. 1986), and in the New World in Mexico (Pijoan
et al. 2007). Though ethnographic claims of cannibalism have been challenged
49
(Arens 1979; Barker et al. 1998; Goldman 1999), archaeological evidence pointing to
this practice prehistorically is difficult to ignore. In all cases, the conclusion of
cannibalism is based on the recognition of damage from corpse processing that is
similar to that found on butchered game animals for consumption as food. I will
explore this topic in more depth.
White defines cannibalism as "conspecific consumption of human tissue"
(1992:9). Early on, accusations of cannibalism were based on ethnographic accounts
gathered by ethnographers, missionaries, and explorers. Often, the informant never
witnessed the practice, but perpetuated the stories or myths based on word-of-mouth
(Arens 1979). However, recognition of this practice using archaeological bone
assemblages removes the subjectivity of the informant and the naiVete of the
ethnographer who is willing to believe much of what an informant reports.
Cannibalism Diagnosis
It is recognized that accusing someone of cannibalism is a serious charge and
thus, the threshold of diagnosing it in the archaeological record is set high (White
1992:8). Today only the most severe occurrences tend to be reported as cannibalism,
whereas assemblages containing a few bones showing cutmarks or a few broken
bones are noted, but are often discounted as anomalies. In the larter half of the 20th
century, research has focused on the recognition and quantification of archaeological
bone modification. Using this new body of knowledge it is now possible to recognize
whether the agents responsible for bone modification are human or non-human. As a
50
result, researchers interested in characterizing skeletal trauma now have more tools at
their disposal.
Turner and Turner (I 999:chapter 2) offer a thorough discussion of skeletal
damage which they interpret as resulting from cannibalism. The following discussion
is abstracted from this source. There are three basic observations that must be made:
I) the damage pattern on the human bones is similar to that of butchered game
animals, 2) the damage appears dissimilar to that caused by animal or environmental
forces, and 3) cannibalized human bone deposits appear dissimilar to deposits
resulting from formal interment. Let us consider each of these separately.
The most striking observation made of butchered human bones is that the
damage frequency and patterns are nearly identical to those observed in
archaeological assemblages of faunal remains. The remains are characterized by a
high degree of longbone fragmentation by spiral fracture, cutmarks, and burning or
charring. Percussion impact sites and anvil abrasions are commonly associated with
the breakage and fragmentation of the bones. Spiral fracture, peeling and splitting of
bones are diagnostic of fresh bone fracture. A low incidence of pot-polish
(smoothing of sharp bone fragments resulting from vigorous cooking in a ceramic
vessel) and the near absence of animal gnawing or chewing are also characteristic of
both butchered human and non-human assemblages.
Animal and environmental damage leave taphonomic signatures that are
different than those left by humans. Animals often gnaw or chew on exposed bones,
an activity that leaves distinctively different damage than cutmarks left by tools,
51
though the two are often confused. The former are characterized by a V-shaped cross
section, while the latter appear V-shaped and may also exhibit ridging within the cut
(Shipman and Rose 1983). Skeletal material that is in sub-surface contexts can
suffer environmental damage from plant roots, which can leave linear marks on bones
and large tree roots, which can penetrate and break skeletal elements. Bone
preservation is also altered by soil content and pH. Calcium carbonate, for instance,
causes bones to crack and the exterior surface to exfoliate. Surface deposited remains
can be altered from the effects of sun-bleaching, which can include severe drying,
cracking, and exfoliation of the outer bone surface.
The deposition and preservation of cannibalized bones is similar to that of
faunal bones that have been cooked. Preservation from this type of preparation and
deposition is markedly different from skeletal material interred while still covered in
flesh. Cannibalized bones are commonly piled or scattered on the surface rather than
formally buried. They are also characterized by an excellent state of preservation,
which is thought to stem from removal of organic materials such as flesh and fat
during cooking. This results in a lack of normal macro- and microorganism activity,
which results in bone leaching by acidic by-products. In sum, cannibalized bones are
most often deposited on the surface and are generally characterized by excellent
preservation, extensive longbone fragmentation and a low incidence ofbuming and
gnawing, whereas non-cannibalized, interred bones are often more poorly preserved,
having suffered organic and chemical degradation resulting from the subsurface
deposition and the activities of soft tissue decomposition.
To summarize, the Casas Grandes skeletal assemblage--especially the non-
interred material-has been grossly understudied. Recent scientific inquiry has
shown that types, patterns, and timing of fractures are useful in reconstructing the
events preceding and following death. Physical and circumstantial evidence should
be treated as paramount in investigations of death whether modem or prehistoric.
52
53
CHAPTER IV
MATERIALS AND METHODS
Sampling
The Casas Grandes Medio period skeletal assemblage is composed of
individuals showing evidence of formal interment (henceforth referred to as interred
skeletons, or IS) and those lacking such evidence (non-interred skeletons, or NIS).
The skeletal material came from 16 excavation units containing unequal numbers of
individuals. The individuals included in this study were chosen as randomly as
possible given the circumstances of their storage. A sampling methodology was
devised to include individuals from all excavated units in equal proportions.
Appendix A contains the list of individuals analyzed and their burial type.
To ensure equal representation from each sub-population (IS, NIS) and from
each unit, the total number of individuals that could be analyzed over the study period
was estimated and this number was converted into a percentage of each sub-
population. These percentages were applied to the separate sub-populations from
each unit and the corresponding number of individuals was calculated for each. A list
of individual burial numbers corresponding to each sub-population was drawn up
using DiPeso and colleagues' (1974:8:359-409) list and these lists were taken to the
storage room for location of representative individuals. Coffin boxes of interest were
visually located and removed for analysis if they fit into one of the sub-populations of
54
interest. The stored coffin boxes were stacked in no particular order and in some
cases individuals listed in the site report were absent from the storage facility. Boxes
contained as few as a single individual and as many as four or five depending on the
size and/or completeness of the individuals contained within them. One hundred
seventy-nine-42 NIS, 137 IS-were evaluated for this study.
Sorting of individual skeletons was at times difficult, if not impossible in
some cases. A passage from Barbara Butler's Ph.D. dissertation explains the source
of much of this difficulty:
"1) The archaeologists tried to excavate complete individuals but
sometimes several individuals were buried together and mixture of
individuals occurred both in the ground and as they were being
removed.
2) In the lab, examples of bones with pathological lesions were
removed from the storage boxes; also maxillae and mandibles were
removed and placed in separate storage areas. Many teeth had been
removed from their sockets (for separate examination). 3) Some
remains were very fragmentary as the result of poor preservation in the
ground and poor handling and preservation in the lab" (Butler
1971:15).
It is important to note that following Butler's analysis, a number of other researchers
examined the skeletons in their research, so an additional amount of mixing and mis-
identification of skeletal elements has occurred. It was not uncommon to find a box
containing multiple individuals-sub-adults and adults alike-but with all the teeth in
one bag marked "Teeth" and one burial number on the bag. Thus, the teeth in
question had to be sorted by chronological age, grouped accordingly, and reunited
with their respective skeletons.
55
Data Collection
The skeletons were analyzed for injury. They were inspected for evidence of
unhealed, healing, and healed fractures, cutmarks, projectile injuries, and burning
(Frayer 1997; Lambert 1997; Ortner and Putschar 1981; Owsley 1994; Sauer 1998;
White 1992). It was expected that if warfare was endemic in the region, individuals
would show a combination of healed and unhealed injuries (Lambert 1997;
Roksandic et al. 2006; Torres-Ruffand Costa Junqueira 2006; Tung 2006).
Description of injuries follows Lovell's (1997) protocol, which takes into account the
location, shape, and type of fracture, as well as any subsequent remodeling or
infection.
Data were collected in the modern town of Casas Grandes, Chihuahua,
Mexico over an eight-week period during the summer of2007. The skeletal
collection is stored in a room at the city's Palacio Municipal (city hall). Analysis was
performed in a similar room next to the storage room. The storage room is not
climate-controlled and during summer months temperatures can typically exceed 90
degrees Fahrenheit with winter temperatures approaching 30 degrees Fahrenheit.
Thus, these conditions are not ideal for the long-term preservation of skeletal
materials, which is reflected in their current state of preservation.
Data were collected during 8-hour days (7 a.m.-4 p.m. with one hour for
lunch), 5 days per week. This schedule allowed for analysis of4 to 7 skeletons per
day depending on their completeness and the amount documentation required for
each. Skeletons that were more complete or showed extensive trauma, pathology, or
56
those requiring several photographs required more time than the more fragmentary
skeletons that showed little to no trauma or pathology.
Data were collected using visual and metrical methods. Visual analysis was
performed under acutely angled secondary light using a lOx hand lens and natural
sunlight when needed to visualize smaller features. Measurements were taken to the
nearest 0.1 mm using a CenTech digital sliding caliper. Digital images were captured
in TIFF format using an Olympus E-500 35 mm DSLR camera with a 17.5-45mm
zoom lens. These images were processed using Adobe Photoshop CS2 software from
which JPEG images were created and stored on gold compact discs. Copies of these
discs were sent to INAH, Chihuahua.
Burial Provenience Data
I separated the NIS into more distinct groupings to look for any patterns or
trends in their deposition. This procedure was similar to DiPeso's coding of the IS.
The scheme was as follows: 2A, floor surface articulated; 2B, floor surface
disarticulated; 2C, fill articulated; 2D, fill disarticulated, and; 2E, no provenience/test
trench.
Sex and Age Determination
The study of archaeological human remains yields information about
mortality and longevity in earlier populations (Lovejoy et al. 1985:1). Inferences
about fertility and crude death rates can be made using this information. Many of the
methods used herein are based on modem reference populations, which allegedly
reduces their reliability when applied to prehistoric populations (Bocquet-Appel and
57
Masset 1982). (The reader is directed to Buikstra and Konigsberg [1985] for further
discussion and rebuttal). Bearing this in mind, the following broad age cohorts were
used: fetus/perinatal «birth), infant (0~3 years), children (3~12 years), adolescents
(12-20 years), young adult (21-35 years), middle adults (35-50 years), and old adults
(50+ years) (Buikstra and Ubelaker 1994:9).
The construction of a paleodemographic profile of the Casas Grandes
population was not a central purpose ofthis study, however, knowledge ofthe
demographic structure of the skeletal sample could have important implications when
justifying warfare or a massacre. Therefore, sex and age were determined for all
individuals when possible.
Sex was determined using the standards described by Buikstra and Ubelaker
(1994). Some osteological traits are sexually dimorphic in humans. Morphological
features of the pelvis and cranium are the best sources for accurately determining sex
from skeletal remains (France 1998). Several researchers (Phenice 1969; Rathbun
and Buikstra 1984:Table III~I; Sutherland and Suchey 1991) report a sex
determination accuracy of90 to 96% using pelvic methods, while the skull returned
accurate determinations of 80-90%. Konigsberg and Hens (1998) reported an
accuracy of 79% when using only the cranium. In many cases the diagnostic portions
of either the cranium or pelvis were missing, forcing me to use only one or the other.
Some additional burials contained neither pelvis nor cranium, in which case I was
forced to make provisional sex determinations based on extreme divergence of the
diameter of the femoral or humeral head from the mean diameter (France 1998).
58
Sexually dimorphic attributes of the pelvic bones are preferred for
determining sex from the skeleton. Differences in expression and scoring of the
pelvic and cranial sexing attributes are summarized in Table 1 and discussed in more
detail below. The subpubic region contains three sexually dimorphic attributes, the
ventral arc, subpubic concavity and ischiopubic ramus ridge. They are scored with
values of 1 (female) to 3 (male). The greater sciatic notch ranges from broad in
females to narrow in males and is scored with values of 1 to 5, respectively. The
preauricular sulcus is thought to appear more commonly in females than males.
Scores range from 0 (absent) to different expressions ranging from 1 to 4 (wide to
narrow). Each aspect should be weighted equally and the total morphological suite
should be used in the determination.
Features of the cranium can be used for sex determination, although they are
slightly less reliable. Buikstra and Ubelaker (1994) illustrate five aspects ofcranial
morphology used to distinguish sex: 1) robusticity of the nuchal crest, 2) size of the
mastoid process, 3) sharpness of the supraorbital margin, 4) prominence of glabella,
and 5) projection of the mental eminence. These are scored on a scale of 1 to 5 with 1
being characteristic of females, 5 being male and 3 being generally ambiguous. All
traits should be weighted equally and considered together.
The head of the humerus and femur were used in some instances to determine
provisional sex. This method is described in detail by France (1998). The head of the
humerus has been found to yield correct sex determinations in archaeological Native
American populations with accuracies between 89 and 95%. Meindl and Russell
59
(1998:379), however, caution that sexing the skeleton based solely on pure size is
often problematic due to the role that the environment played in growth in prehistory.
Therefore, this method is used as a last resort since often times these portions of the
skeleton are preserved when other sex-determining elements are missing.
Table 1. Skeletal sexing attributes (compiled from Buikstra and Ubelaker [1994]
and France r1998]).
Score (except preauricular sulcus)
1 3,4, or 5
Range of Variation
Female ~---------------------------------------------------7 Male
Pelvis
Subpubic angle
Ventral arc
Subpubic concavity
Ischiopubic ramus ridge
Greater sciatic notch
Preauricular sulcus
More than 90 degrees
Well-defined
Concave
Narrow, crestlike ridge
Wide
Present
Less than 90 degrees
No ridge (not well-defined)
Convex
Broad and flat
Narrow (fits snugly around your thumb)
Absent
Cranial
Nuchal crest
Mastoid process
Supraorbital margin
Prominence ofglabella
Mental eminence
Smooth, no bony projections
Short, narrow
Sharp
Smooth, no projection
Little to no projection, single point
Bony ledge or hook
Long, robust
Dull, blunted
Prominent, rounded
Massive, bilobed
Post-cranial
Humerus (transverse)
(vertical)
Femur (max. diameter)
Smaller
Smaller
Smaller
39.7 mm
42.4mm
44.6mm
Larger
Larger
Larger
The method requires measurement of either the vertical or transverse diameter
of the head of the humerus, which is compared to published cutoff points. For this
study, the cutoff points of the Arikara and Pecos Pueblo populations (as described by
France 1998) were averaged, which produced a cutoffpoint of39.7 mm for the
transverse diameter and 42.4 mm for the vertical diameter. Thus, a humerus head
60
with a transverse or vertical diameter smaller than its respective cutoff point is most
likely from a female; one whose diameter measures larger is most likely from a male.
Sex determination from the maximum diameter of the femoral head is done in a
similar manner using a cutoff figure of 44.6 mm. This method yields an accurate
determination 88.7% of the time and is based on a central California Native American
population (France 1998).
Age at death in archaeological material can be estimated from both dental and
skeletal tissues. In subadults, the sequence of dental development is the preferred
indicator (Baker et al. 2005:157), as it best approximates the true age, yielding ages
with ranges of variation of as little as six months (Ubelaker 1989). This is due to the
more regular formation and eruption of teeth as compared to the more variable
epiphyseal fusion times of the bones.
The dentition was the primary source for subadult age estimation, using
formation and eruption sequences described in Buikstra and Ubelaker (1994). The
skeleton was used as the secondary source of age estimation. This was done using
epiphyseal closure sequences as described by Baker and colleagues (2005), Buikstra
and Ubelaker (1994), and White and Folkens (2005). Age from subadult skeletal
tissues is estimated by observing the appearance and fusion ofthe 450 skeletal
elements present at birth as they develop into the 206 bones present in adults (Baker
et al. 2005:6). Growth proceeds from the ossification centers of the bones. Once
growth is completed, the ends of the bone portions fuse at cartilaginous plates, known
as epiphyseal plates, which ossifY, uniting the pieces into a single bone. The timing
61
of these ossifications is more or less regular and their rate can be used to approximate
age at death.
Age of adult skeletons was estimated by observing degenerative changes in
the teeth and skeleton. Age was estimated using morphology of the symphyseal
surface of the os pubis (Brooks and Suchey 1990; Buikstra and Ubelaker 1994) and
auricular surface ofthe ilium (Lovejoy et al. 1985; Buikstra and Ubelaker 1994),
dental attrition of the anterior teeth (Smith 1984) and molars (Scott 1979), and cranial
suture closure (Meindl and Lovejoy 1985). Sex and age of many of the NIS were
unable to be determined as they lacked the correct elements required. All of these
indicators were used in as many instances as possible, though most age estimations
were based on less than the full suite due to skeletal damage and missing elements.
Age-related changes ofthe sympheseal surface of the pubis are considered the
most reliable criteria available for approximating age at death in adults (Buikstra and
Ubelaker 1994:21). However, it is often under-utilized since this portion of the
skeleton does not survive well archaeologically. A method for evaluating the
auricular surface of the ilium has been developed due its better survival
archaeologically, though this method is more complex and difficult to score. In both
cases, the presence or morphology of key features must be evaluated.
The pubic symphysis is observed for the following features: ridge-and-furrow
system, dorsal margin, dorsal platform, ventral rampart, ossific nodules, rim, and
delimited extremities. The presence and morphology of these features are compared
to six phase-by-phase descriptions by Brooks and Suchey (1990), which divides the
62
morphological continuum. Each of the phases correlates with an age range. Sex of
the individual must first be determined as this method takes into consideration the
differences in age-related changes between the sexes. The precision is greater for
males than females, which may be due to greater variation in pubic morphology of
females due to childbirth or the three-fold difference in sample size of males over
females.
Age-related assessment of the auricular surface of the ilium proceeds in a
similar manner, though morphological attributes of different regions of the auricular
surface are considered. Four regions of the surface (apex, superior and inferior
demifaces, retroarticular area) are graded for relief, granularity, density and porosity.
These qualitative assessments are compared to phase descriptions as in the pubic
method, and each phase correlates with an age range. Both pelvic aging methods are
described in more detail in Buikstra and Ubelaker (1994) and White and Folkens
(2005).
Methods ofData Analysis
Most data were compared using the G-test of independence. The G-test (or
log-likelihood ratio) is preferred to the chi-square as it better approximates the
theoretical chi-square distribution when a cell has a larger than expected absolute
value (Sokal and Rohlf 1994). The test was performed using DOS-based computer
software published by James Mallet (2006) of the Galton Laboratory, Department of
Genetics, Evolution and Environment, University College, London. The test reports
63
two statistics, G and x2, and a P value, which represents the probability that the
experimental results are due to chance.
Fisher's exact test was used to compare trauma frequencies between the sexes,
the IS and NIS, and between Paquime and other sites. This statistical test is applied
to evaluate whether the differences in sample proportions of binary data are due to
chance (Tung 2007). Fisher's is useful for small sample sizes and reports a P value,
which represents the probability that randomization alone is responsible for the
disparity observed between two proportions (Ramsey and Schafer 1997:548-550).
Pair-wise differences were the subject of interest of the element representation
comparison. The Wilcoxon signed-rank test was used to compare paired frequencies
of skeletal elements. Though some useful information about the magnitudes of pair-
wise differences is excluded, the Wilcoxon mitigates the effects of outliers (Ramsey
and Schafer 1997:96).
Dating
Corticallongbone samples from 19 individuals were collected for use in
chronometric dating. The samples were from both IS and NIS. These were collected
individually and immediately placed in separate plastic Ziploc bags to avoid mixing
and contamination. Portions of these samples were sent to the University of Notre
Dame Department of Anthropology for analysis of fluoride ion concentration. Five
gram samples were sent to the University of Arizona NSF Laboratory and Geochron
Laboratories for AMS dating.
64
Fluoride ion analysis measures the amount of fluoride taken up by buried
bones from ground water (Schurr 1989:265). It works well for bones that were
deposited less than 1,500 years ago and over a short period of time such as a few
hundred years (Haddy and Hanson 1982). The method is very sensitive and can
statistically distinguish samples with fluoride concentration differences of 0.024%,
which can represent a time span of 10 to 20 years (Schurr 2007). It is helpful when
sorting individual samples or sub-assemblages chronologically. Samples of interest
can then be selected for the more costly AMS dating (Aitken 1990:219). This method
provided key evidence in debunking the Piltdown forgery by showing the skull and
mandible of the purported "missing link" were of different ages (Spencer 1990).
AMS yields absolute (calendrical) dates. AMS dating has its roots in the post-
WWII discovery that the age of materials containing radioactive elements could be
determined by measuring the amount of decay of the elements (Price and Feinman
1993: 13; Taylor 1997:66-67). Plants and animals ingest radioactive carbon cr 4C)
throughout their lives. When an animal's carbon intake ceases (the animal dies), the
process of decay of the radioactive carbon begins (Price and Feinman 1993:13-14).
An approximate time since death is rendered by knowing the rate at which 14C decays
(the half-life) and measuring the amount of radioactive decay that has occurred.
Five milligram samples of adult corticallongbone were weighed on separate
papers on a triple-beam balance. These were submitted to the University of Notre
Dame for fluoride ion analysis, a process that took approximately one month. Once
65
results were obtained from fluoride ion analysis, samples from five individuals were
submitted for AMS dating.
The University of Arizona NSF Radiocarbon Laboratory was the primary
facility used for dating, and sample size for all AMS dates was 5 grams. The oldest
and youngest NIS and IS samples (those with the greatest and least concentrations of
fluoride) were chosen to bracket the lower and upper temporal bounds of both groups.
A fifth individual from the middle of the NIS group was dated by AMS as its fluoride
measurement was similar to one other NIS sample and two IS samples. Thus, a
single AMS date from this sample would yield an approximate date for three others.
Samples of the upper and lower NIS were also sent Geochron Laboratories, which
was used ~s a r~ference ~h~ck against the University of Arizona Laboratory. The
sarrlples were' ptetreated by the laboratotiesusing 'it'in'o,Htled version'ofLongin; s
(1971) technique for eliminating contaminating sources of carbon and isolating bone
collagen. The AMS dates were originally calibrated with both CALIB 5.0.1 and
OxCal 4.1.1 using the JntCa104 data set curve for northern hemisphere non-marine
.. '. ~ :.'.
~f!.mpl~~Cl}ronk Ramsey 1995; 2001; Reimer et al. 2004; Stuiver and Reimer 1993).
The two programs gave similar results, though the OxCal output was chosen over
CALIB for its greater graphic capabilities.
66
CHAPTER V
RESULTS
Sample Population Demographics
The results of age and sex detennination are presented in Figures 7 and 8.
Reproductive theory dictates that the sexes are represented approximately equally in
primates (Dlmbar 1984:196). My comparison of frequencies of individuals ofknown
sex shows a 20% female bias (refer to Table 2). DiPeso et al. (1974:8:359) reported a
similar figure for the IS adult population, which is slightly higher than the 16%
reported here, though his frequency of unknown sex was only 13.9% compared to
43% here. The source of the some of the discrepancy may be due to their use ofthe
"? Adult 18+" age category. According to Butler (1971), age detennination was
made almost exclusively from dental eruption, toothwear, and cranial suture closure.
In cases of young adults or juveniles where there were no ageable teeth, age was
detennined from the spheno-occipital synchondrosis, which made the individual
adolescent or adult. Thus, the precision and reliability of these ages, based on a
single suture should be questioned. Due to inherent human variation, the possibility
remains that some individuals in this category were in their mid-teens, in which case
sex detennination would be questionable. According to chance, the sexes of this
67
70
60
50
80 ~------------------------------------------I
I
~-------~------~I::::::i--------~-----------I
- --5IGunknO";-L I
_____ II Female _ _ 1
.Mal~ I
---------------------1
I
:
20 ------1::::: :1-------
30 ----------------------------
10 .-----
"E
~ 40 ----------------•.--------------
()
Ago Group
Figure 7. Demographics of the study population
601 -
50 r--------------------
40
~g 30· -----
o
20 +----------------------
II
1
~-~-NIS--_J
~~
----- 1
I
--·----1
Figure 8. Age distribution of NIS compared to IS.
68
group should be represented approximately equally. However, the numbers of
individuals ofknown sex from this age category (not considering those for whom sex
was unknown) gave totals of9 males and 29 females, a result highly divergent from
chance. Moreover, 6 juveniles between the ages of 13 and 19 were sexed-all of
them female. Hence, 80 percent of the 44 individuals who were either in their teens
or about 18 years of age were sexed as female. This is not surprising given that the
human skeleton looks morphologically female until after puberty. The point ofthis
discussion is that by chance, approximately one half of the 44 individuals should be
female and the other half male.
Table 2. Total counts and frequencies of sex by sub-population.
Males Females Unknown
Population Males (%) Females (%) Unknown (%) Total
NIS + IS (total) 42 23 64 36 73 41 179
NIS + IS (known sex) 42 40 64 60 106
NIS 13 31 15 36 14 33 42
IS 28 20 49 36 59 43 137
IS (DiPeso et al. 1974) 86 33.2 137 52.9 36 13.9 259
IS (Butler 1971) 106 21.7 153 31 234 47.5 493
To test this possibility, the 44 individuals were divided into equal halves by
sex. This division both increased the males and reduced the females by 13. This
decreased DiPeso's IS frequency difference from 19.7% to 9.8%. Given this small
difference, and that neither reported sex frequency accounts for more than one half
the total population, the group of unknown sex could make up for the disparity. This
leaves open the possibility that some of the female sex bias is introduced rather than
real. The sex and age frequencies from this study are not significantly different from
69
those reported by Butler (1971 :21) for Medio period skeletons (G=.66, d.f.=2, P> .05,
G=7.53, d.f.= 5, P> .05, respectively) (Table 2).
Comparison ofNIS and IS demographics showed some disparity, with age
distributions showing significant difference (G=32.04, d.f.=7, P < .01). There were
no individuals under the age ofthree or over the age of 35 in the NIS (see Figure 8).
If the IS demographics are considered to represent the expected distribution of the
general population at any given time of site occupation, the age structure of the NIS
departs from the distribution. Sex distribution was not significantly different between
the IS and NIS groups (G=.30, d.f.=2, P> .05).
Non-Interred Skeleton Burial Provenience
The NIS burial provenience data show that 15% ofthe NIS were articulated,
and 83% were fragmentary and not articulated (see Table 3). Forty-one percent were
deposited on floor surfaces, 57% were deposited above floor surfaces and mixed with
fill. The only discernable pattern was that the articulated skeletons buried in fill (type
2C; 12 individuals) were distributed almost evenly among units 8, 14, 15, and 16 (one
was recovered from unit 4). In short, what DiPeso and subsequent authors considered
one burial group is actually made up of four unique sub-groups. The differences seen
here will have consequences for the interpretation of the NIS group representing a
massacre.
Antemortem and Perimortem Fractures
Evidence of skeletal trauma was observed, but in much lower frequencies than
expected for an assemblage representing long-standing warfare. Fourteen individuals
70
Table 3. Provenience coding and count of NIS skeletal material, which includes
the two type 9 burials (DiPeso's "accidental death" is a manner of death
designation, not a burial type).
Count
NIS Burial (percent
Type of total)
2A 8 (6)
2B 45 (35)
2C 12 (9)
2D 62 (48)
__2E l (1)
(7%) showed evidence of ante- or perimortem fracture. A total of 15 fractures were
recorded (one individual had two fractures). Nine individuals had antemortem
fractures, four had perimortem fractures, and one had both. Tables 4 and 5 present
this information in more detail. Frequencies of longbone fracture compared to the
total number of skeletons analyzed were as follows: clavicle 1.7%, radius 1%, ulna
1.8%, fibula 3.5%. Ten of these fractures were found on the IS, four were found on
the NIS. Two cases of antemortem forearm fracture were noted, which could be
parry fractures or simply the result ofaccidents (Figure 9). One male survived a
potentially lethal blow to the right facial region that fractured the right frontal and
nasal bones resulting in orbital and nasal defoffility (Figure 10). A single juvenile
NIS (burial 29-8) suffered a perimortem blow to the right parietal, which left an
opening into the skull (Figure 11). This was one of two NIS where interpersonal
violence may have caused the injury, though it could be an artifact of postmortem
processing of which there was further evidence. A second NIS (buriaI30A-l) had a
Table 4. Detail of skeletal trauma separated by timing.
Skeleton
Number Trauma
Antemortem
20-1 Left distal ulna (NIS)
30-1 Right distal fibula (IS)
35-6 Right distal ulna, radius (crush?) (IS)
5-8 Right clavicle (IS)
24-8 Right clavicle (IS)
1-11 Right distal fibula (NIS)
34-12 Left 5th metatarsal (IS)
23-14 Left distal fibula (NIS)
1-21 Left metatarsal (IS)
Perimortem
30A-l Right frontal (IS)
29-8 Right parietal (NIS)
15-12 Left distal fibula (IS)
13-13 Left parietal (IS)
39-14 Left parietal (IS)
Table 5. Tally of fractures by timing and location (total of 14 individuals).
Location Antemortem Perimortem
Ulna/radius 2
Fibula 3
Metatarsal 2
Clavicle 2
Skull 1 4
Total 10 5
perimortem depressed cranial fracture, which may also have been an artifact of
postmortem processing, as the skeleton was very fragmented (Figure 12). No
projectile injuries were detected in any of the material. Statistical comparison
showed no significant difference between the IS and NIS groups (G=.32, d.f. 1, P
>.05).
71
72
Peri-IPostmortem Fractures
Several individuals showed perimortem modification that occurred soon after
death and often after soft tissue had been removed from the bones (see section on
bone splitting, above). I have termed this as postmortem modification (PM).
Postmortem modification was encountered on the remains of 36 individuals
composed of both IS and NIS. Twenty-two (52 %) of the 42 NIS showed evidence of
postmortem modification, whereas 14 (10 %) of the 137 IS showed similar damage.
This PM group was characterized as highly fragmented and incomplete, although two
individuals were approximately 40 to 60% complete. The age distribution of this
assemblage is shown in Figure 13. The damage occurred to the bones soon after
Figure 9. Normal ulna (left), ulna with healed fracture (right) (burial 20-1).
73
Figure 10. Old adult male with healed cranial and facial fractures (burial 20-8).
Figure 11. Portion of child's parietal with hinge fracture (burial 29-8).
74
Figure 12. Portion of frontal bone showing unhealed depressed fracture (arrow)
(buriaI30A-1).
------
I
-------------i
14- ---------------
16 - -----------------------------------1
I
-----------]
I
12 ------------
10
0-3 3-12 yr 12-20 21-35
Age Group
50+ Unknown
Figure 13, Age distribution of group with postmortem modification,
75
death as the breaks were acute- angled, spiral, concoidal, and often had adhering bone
fragments. Additionally, peeling was found on some specimens, which also indicates
fresh bone breakage. Other damage included cutmarks, charring/burning, percussion
sites, and anvil abrasions. These PM remains were characterized by an excellent
degree of preservation, both absolutely and relatively when compared to remains
lacking this damage.
The PM group was characterized by extreme bone fragmentation, especially
the crania and longbones. Semicircular percussion sites and striations indicative of
abrasions caused by an anvil were numerous on longbone fragments. Many
percussion sites had adhering bone fragments, which indicate the bones were broken
while fresh. Examples of this damage are seen in Figure 14. A striking example of
bone fracture was found on both os coxae ofa single individual (9A-15) where the
iliac blades were removed. Figure 15 shows these bones after re-assembly. Table 6
summarizes the types of skeletal modification by individual.
Figure 14. Longbone fragment with various types ofperi-/postmortem damage.
76
Figure 15. Pelvic innominates of a single individual (9A-15) showing
postmortem chop marks (arrows) for removal of iliac blades.
Cutmarks were found on fragments representing nearly every element of the
skeleton. The bones of the skull were most frequently affected. Specifically, the area
of the frontal bone superior to glabella, the supramastoid crests, and the nuchal line
were the most common sites of damage. Cutmarks to the anterior edge of the
coronoid process occurred on two specimens. Cutmarks on longbones were found
most frequently on the articular areas near the ends of the bones, which are likely
indicators of dismemberment (White 1992:267). A single sternal body (9B-15)
showed multiple vertical cutmarks running its entire length. Evidence consistent with
decapitation (Haverkort and Lubell 1999: 157) came from a single individual (9-11)
with two cutmarks to the articular surface of the odontoid process (Figure 16). A
single right scaphoid bone (1-11) showed four distinct cutmarks.
Table 6. Summary of postmortem skeletal modification.
77
Unit BurialNumber
Burial
Type
Percent of
Skeleton
Recovered
Sex CM BU SC BT DC PO
14 N 5 U x x x
20 N <5 M x
30 I 5 U x
19A N 10 U x x
19C N <5 F x x
21A N 15 F x
30D I 5 U x x
6 4 N <5 U x x
17 I 20 U x
24 N <5 U x x
31 I 40 U x x
5A N 15 M x x x
5B N <5 U x x
8 44 I 25 M x x
11 1 N 15 M x x x
3 I 10 F x
8 N 10 M x x
9 N 20 F x x x
18 N <5 F x x
12 31 N 10 F x
34 I 60 M x
13 16 I <5 F x
19 I 15 F x
14 23 N <5 U x
25 N 5 F x x
27 N 5 U x
28 N 5 U x x
15 2 N 30 M x x
9A I 20 F x x
9B I 10 M x x
19 8 N <5 U x x
CP 21A I <5 U x x
21B I <5 F x x
21D I <5 F x
I=IS, N=NIS, F=female, M=male, U=unknown, CM=cutmarks, BU=burning, SC=scalping, BT=blunt-
force trauma, DC=decapitation, PO=pot-polish.
78
Figure 16. Second cervical vertebra with cutmarks on the odontoid process.
Complete longbones were rarely encountered in any of the NIS. Skull
fragments from 13 individuals showed evidence of scalping. Figures 17 and 18
illustrate several types of this damage. Table 7 compares this group of 36 individuals
with Southwest sites showing similar damage used as evidence of cannibalism.
Ten individuals showed evidence of burning/charring with color ranging from
light tan to black in color. Burning was encountered on seven skulls, with a
consistent pattern of thermal coloration concentrated on the endocranial surface of the
calotte (Figure 19), though one mandible (2lA-l) was burned. Three talus bones
(two right, one left), one rib fragment, one fibula and one unspecified longbone
fragment also showed evidence of burning.
Figure 17. Longbone assemblage from a single individual (9A-15.
Figure 18. Partial skull showing multiple diagonal cutmarks running upper
right to lower left across the frontal bone (burial 8-11).
79
80
Table 7. Comparison of Paquime with Southwest sites showing evidence of
cannibalism.
Cowboy Castle Polacca
Mancos Wash Rock Wash Bumt
(5MTUMR- (5MTlOOl (5MT- (NA8502) Mesa
Paquime 2346)a O)b 1825)" d (LA4528)e
surface context X# X X X X
fragmentation X X X X X X
anvil abrasions X X X X
concoidal breaks X X X X X
cutmarks X X X X X X
few verts, hlf bones X X X X X
burning (low) X X X X X X
pot polish (low) X X X
well-preserved X X X
gnawing absent X X X X X X
charred crania X X X X X
#floor surface and in fill
aWhite 1992; bBillman et al. 2000; cKuckelman et al. 2002; dTurner and Morris 1970; eFlinn et al. 1976.
Figure 19. Charred and fragmented skull. Note that charring is restricted to the
outer crown of the skull.
81
Three examples of pot-polish were noted on a longbone fragment, a clavicle, and a rib
fragment. The blunted bone edges suggesting pot polish were visualized under 10-
power magnification, though an often shiny appearance was seen with the unaided
eye. Rodent gnawing was noted on only two longbone fragments.
Statistical comparison of individuals of known sex showed no significant
difference in the occurrence of postmortem modification between the sexes of either
the IS (P= 0.221, Fisher's exact; N=80) or the NIS (P=0.665, Fisher's exact; N=22).
Thus, neither sex from either provenience was more likely to show this modification,
though IS females were three times more likely than their male counterparts. Inter-
site comparison of Paquime (N= 18) and Burnt Mesa LA 4528 (N=6) (Flinn et al.
1976) showed no significant difference for the occurrence of this modification
(P=0.169, Fisher's exact). Difference in the frequency of postmortem damage
between the IS and NIS was highly significant (G = 24.19, d.f. 1, P < .001).
Spatial Distribution
Postmortem modified remains were encountered in every fully excavated unit
(see Figure 20), though not all NIS showed evidence of PM. The bulk of remains
were deposited in units 1,6,11, and 14. Postmortem modified remains were found in
units 13, 15, 19, and Central Plaza, but in much lower frequencies. Non-interred
skeletons were found in units 8, 12, 15, and 19. This could imply that units 1,6, 11,
and 14 were abandoned earlier than 8, 12, 19, and Central Plaza. However, units 8,
16, and 19 were not completely excavated, so they may still contain more burials.
Curiously, the modified remains found in unit 13 and the Central Plaza were only
82
from sub-floor interments. This may imply a ritual activity unrelated to the non-
interred remains found in the rooms as DiPeso (1974:8:336) initially asserted.
Statistical comparison of the spatial distribution of individuals with postmortem
modification showed no significant difference between excavated units (G = 15.04,
d.f. 9, P> .05).
7 ,/ ::::
6 ,/ .:.
4 ---
':':'~---,
--------i:r
~---=-----------""
11 126
:':::
,-,---'='~'---""T-'=,--"'=,-, -"=,,-""'S-,-,-,~- I ,-----r-- i i ,/
13 14 15 16 19 20 21 22 23 CP
Unit Number
2 "
Figure 20. Distribution of postmortem modified remains by burial provenience
and unit.
Skeletal Element Representation
The PM skeletons were highly fragmented and incomplete. Turner and
Turner (1999) maintain that cannibalized assemblages are characterized by a paucity
of ribs and vertebrae as compared to longbones. Though this could be the result of
processing related to cannibalism, other taphonomic processes can account for these
discrepancies (see page 46). To determine if this pattern held true for Paquime, bone
fragments were counted and separated by skeletal element or into groups. These
83
counts were converted into frequencies of total bones recovered. They were
compared to the frequencies of occurrence of whole bones of a complete human
skeleton. Turner and Turner's (1999) assertion holds true for Paquime. Figure 21
illustrates this with the arms and legs considered separately. The pattern becomes
more apparent in Figure 22 when all identifiable longbone fragments are pooled.
The PM group was characterized by much higher than expected frequencies of
cranial bones and longbones, while frequencies ofvertebrae, and hand and foot bones
were approximately one-half and one-quarter the frequencies expected, respectively
(refer to Table 7). Frequencies of axial elements and ribs were nearly identical to
those expected. The Wilcoxon signed-rank test showed no statistical differences
between the paired element frequencies (Z=.262; two-sided p=.79). However, the G-
test did show a significant difference between the observed and expected frequencies
for the entire skeleton (G=59.43, d.f. 5, P < .01). Though I realize I am comparing
fragmented archaeological bone counts to whole bone counts from a complete
skeleton, the frequencies of elements should be similar.
84
Ribs
---------------~-----
AnnsAxialCranial
O--¥-----,---------,-------,-----I,--------',-----,----~
Legs Misc.Longbone Vertebrae
Skeletal Element
800
200
400
1000
1200-
c:8 600-
Figure 21. Bone fragment count (arms and legs considered separately).
1:
"oo
Cranial Axial Longbones Vertebrae
Skeletal Element
Ribs Hands/Feet
Figure 22. Bone fragment count (alliongbones pooled).
85
Table 8. Comparison of skeletal element frequencies (%) between a laboratory
skeleton (expected) and Paquime (observed) remains.
Difference
14
1.4
31
-5.2
0.2
-41.1
Paquime
25
5.9
37
6.8
12.2
12.7
Skeleton____L.--__
11
4.5
6
12
12
53.8
Skeletal
Element/Group
cranial
axial
longbones
vertebrae
ribs
hands/feet
In summary, 52 percent of the NIS and 10 percent of the IS showed
postmortem modification and/or burning related to cannibalism. The damage
included: extensive fracturing and splitting of longbones; chipping and abrasions;
cutmarks to the cranium and postcranial elements, with focus on the articular areas
and major muscle attachment sites of the longbones; burning/charring mainly focused
on the calotte, and; one case of probable decapitation. The group was characterized
by an under-representation of vertebrae, and hand and foot bones, and the remains
were distributed throughout all excavated units.
Fluorine Dates
Results from fluoride ion concentration analysis represent a skeletal
assemblage that was deposited continuously through time (see Figure 23). The
distribution of the NIS and IS are similar when plotted against each other (Figure 24).
No distinct clustering is seen that would indicate a large, single skeletal deposition
event. Appendix II lists this information in more detail.
86
f
!
f I
I
0.8 j---------------
I
0.61------·----------------------------
, • !
1.2 i
I
I
I1r
I
,
t-~11'
O.2r-+- --
01
Figure 23. Fluoride ion concentration distribution. Dots represent mean
fluoride concentration from three replicate measurements; lines represent
standard deviations. Chronologically younger samples are at the bottom, older
are at the top.
------------------------------:;7«---.---
-/---/----------------- i:::~sl,
0.4 r--------="'--~,'------------------
0_8 -------------+-------;1'-
0.2
Figure 24. Comparison of the distribution of fluoride concentration between
populations.
87
AMS Radiocarbon Dates
Four of the five samples submitted for analysis (17-6, 19-13, 19A-1, and 27-
14) returned AMS radiocarbon dates. One sample (31-12) was not analyzed as it
produced no collagen. The results, with their respective fluoride results are shown in
Table 9. The calibrated AMS dates are shown in Figure 25. It is worth noting that
the average percentages of fluoride in the samples are consistent with their AMS
dates. Table 9 indicates that samples 19A-1 and 17-6 exhibit highly overlapping age
ranges. These two samples appear as shadows of each other on the OxCal plot in·
Figure 25.
Table 9. Summary of AMS and fluoride data.
Average Mean Cal 95.4% (2cr)
percent 14C age yr age yr cal age
Sample Burial type fluoride RP. A.D. range A.D.
19-13 IS .949 689 ± 42 1325 1255-1395
~7-14 NIS 1.076 650 ± 40 1338 1278-1398
19A-1 NIS .255 570 ±40 1364 1298-1429
17-6 IS .198 567 ± 42 1365 1299-1431
In summary, the results show skeletal evidence of antemortem, perimortem,
and postmortem trauma. Division of the type 2 (NIS) burials into four burial types
allows for a better examination of the NIS burial provenience. Fluorine and AMS
dates demonstrate a continuity of deposition of bodies during the recognized
boundaries of the Medio period. These data will be instrumental in determining
circumstances surrounding the deaths of the individuals studied.
Q..
ca
r.:
o
-e
m
'"o
D
f,J
~
700 'R_Dateii9-~3 (Ai'-~'-'------i-
! i
-·-·1--------·---,
! .
500------!-------;------
I •
400 ~.
Calibrated date (caIAD)
88
Figure 25. Raw AMS dates plotted against their calibrated dates. Diagonal
curve is the IntCal.04 atmospheric radiation calibration curve. Peaks above
date ranges are probabilities. Note tremendous overlap of 19A-1 and 17-6
between 500 and 600 B.P. (A) = date from University of Arizona Laboratory,
(G) =date from Geochron Laboratory.
89
CHAPTER VI
EVALUATION OF HYPOTHESES
This chapter evaluates the three hypotheses stated in Chapter I. The primary
line of evidence used is the osteological data. Additional evidence such as
chronometric dates and burial context will be used in evaluation ofhypotheses 2 and
3. When assessing whether violence was responsible for creating archaeological
assemblages, frequencies and patterns of injury may be as important as the individual
traumatic lesions themselves. Table 10 shows the patterns of injury generally
Table 10. Comparison of skeletal damage and preservation associated with
different types of corpse processing.
Mortuary
Mass- process- Witch Cannibal- Scaveng-
Warfare acre Sacrifice ing killing ism ing
Fracture
cranial x x x x
post-cranial x x x x x
chewing x x
gnawing x x
Cutmarks
scalping x x
disarticula-
tion x x x x x
defleshing x x x x
Other
burning x x
well-
preserved x
90
observed on skeletal assemblages resulting from each event. The Paquime skeletal
material will be evaluated against each category to identify similarities and
differences.
Hypothesis 1: Intergroup Violence Was Endemic in the Casas Grandes Region
Hypothesis 1 is evaluated by observation of skeletal trauma and comparing
trauma types and frequencies to other sites where warfare has been confirmed. Low
frequencies of ante- and perimortem skeletal trauma were discovered at Paquime.
The frequency of trauma at Paquime is similar to those from Mohenjo-daro (Kennedy
1984) and Altar de Sacrificios (Saul 1972), two sites where neither warfare nor
massacre have been recorded (see Figure 26). These three sites contrast greatly with
the Southwest sites of Sand Canyon Pueblo and Castle Rock Pueblo where massacres
and cannibalism occurred, the Great Plains sites of Larson and Crow Creek where
massacres occurred, and the California Channel Islands where warfare was endemic.
Though a small percentage of the population suffered ante- and perimortem trauma,
there were no patterned injuries (e.g. multiple healed depressed cranial fractures,
parry fractures coupled with cranial fractures on the same individual) that indicate
warfare was rampant in the Paquime region. It is equally likely that these injuries
resulted from accidents. This pattern differs from the warfare pattern seen in Table
10 where fractures and scalping are common.
DiPeso supported his warfare theory with the paucity of male burials at
Paquime, believing that warriors were likely lost on distant battlefields and/or buried
91
25 --~------~-----~---------"-------"--------"----~-
20
15
~
,.,
u
c:
":J
.,.
!!
lL
10 "
o
Paquime Altar de Sacrificios Castle Rock Pueblo
Site
Sand Canyon Pueblo Channel Islands
------------------------------j
a.
100
90
80
70
60
~
,.,
u
c: 50
":J
.,.
!
lL
40
30
20
10
0
Paquime
b.
Mohenjo Daro Sand Canyon Pueblo Castle Rock Pueblo
S~e
Larson Crow Creek
Figure 26. Comparison of a). antemortem, and b). perimortem trauma between
Paquime and other selected sites.
92
in a special warrior cemetery outside the city. Though this could explain part ofthe
female bias in the sex ratio, it does not explain the virtual absence ofmale warfare
survivors. Ifwarfare was necessary to maintain Paquime's control over its hinterland,
a sizeable army would have been needed, and at least some of these men would have
survived one or moreinjuries incurred in war. The skeletal record does not reflect this,
as healed injuries preserved after death were infrequent. Thus, the osteological data
do not support the hypothesis that warfare was endemic to the Casas Grandes region.
Hypothesis 2: The Final Population ofPaquime Was Massacred
Hypothesis 2 is evaluated by comparing trauma frequencies ofthe IS and NIS.
I expected the frequency of trauma would be significantly higher in the NIS.
Additionally, the demographics of the NIS should be similar to those ofthe city's
general population. Fluoride and AMS dating, and burial data were used in support
as well. The frequency of injury was nearly the same between the NIS and IS
(IS=7%, NIS=lO%) and only one perimortem fracture was found on a NIS.
Comparison of trauma frequencies showed no significant difference (G=.32, d.f. 1, P
>.05).
When compared to Table 10, the NIS show similarities to massacre sites in
some categories. Though the NIS number is over 127 (certainly large enough for a
massacre), they show dissimilarities with remains from sites such as Crow Creek and
Larson. First, the skeletons are incomplete, not articulated, and highly fragmented.
Second, the remains lack evidence of carnivore gnawing, a sign of animal activity on
exposed corpses. Third, very few show injuries related to the cause of death. The
93
NIS did show consistencies with Crow Creek and Larson for evidence of
disarticulation (much higher at Paquime), scalping, and defleshing, and limited cases
of burning. However, the NIS were well-preserved, and much more fragmentary and
scattered than the assemblages at either Crow Creek or Larson.
The results from both dating methods also argue against a massacre. The
relative dating sequence provided by the fluoride analysis shows a pattern of
continuous deposition for both the IS and the NIS. If the NIS are the remains of a
massacre, they should cluster near the point oflowest fluoride concentration (a
horizontal line near the bottom ofthe graph), while the IS values are continuous
throughout the Medio period (a positively-sloped diagonal line) (refer to Figures 23-
25 above). The AMS dates also show the IS and NIS to be contemporaneous, and to
have been deposited during the Medio period.
A similar pattern of burial has been noted for the site of Galaz Ruin in
southwest New Mexico. Anyon and LeBlanc (1984:174) note that at least 25% of
burials from Galaz Ruin were extramural (buried outside structures or in room fill)
much like the NIS at Paquime. My statistical analysis of the two Galaz burial groups
showed no significant difference between the Galaz age distributions (G=8.77 d.f. 5,
P >.05), thus, the extramural burial demographics were similar to a normal Galaz
population. There was a slight tendency to bury people either in or in the fill of or
adjacent to rooms suspected as being ceremonial or public in nature (Anyon and
LeBlanc 1984:182).
94
Fifty burials were recovered from post-abandonment fill in Structure 42A, and
have fewer than average grave goods in them (182). The authors suggest that the fill
in this room would have been soft and a prime place to dig graves (182). Thus, there
are numerous similarities between the Paquime NIS and the Galaz extramural burials,
yet LeBlanc (1999:233) sees massacre as the only event that could have produced the
Paquirne NIS. Though I do not believe the Galaz burials to be massacre victims, the
similarities among these distributions make the Galaz extramural burial group a better
candidate for a massacre than the NIS from Paquime.
The burial data from the NIS do not support their being the remnants of a
massacre. This would have been apparent long ago if their distinctiveness had not
been obscured by their designation as a single burial type. In fact, the type 2 NIS
actually occupy four different burial types. My types 2A and 2C indicate that some
bodies were buried whole, while bodies in burial types 2B and 2D were buried in
pieces and possibly strewn through different rooms. Some bodies were buried in the
fill of abandoned rooms, some were placed on the floor surfaces and covered in fill.
Both dating methods argue that these bodies were interred throughout the occupation
of the city, which is consistent with the burials taking place as different rooms were
abandoned. Hence, a final massacre is not supported.
Hypothesis 3: Human Sacrifice and Cannibalism Were Practiced at Paquime
Hypothesis 3 is evaluated by comparing the types and patterns of trauma
found at Paquime with those from sites where human sacrifice and cannibalism have
been documented. The 1958-1961 excavations recovered ten individuals that DiPeso
95
and colleagues believed were human sacrifices (DiPeso 1974:2:637; DiPeso et al.
1974:8:361). The notion that some of the individuals were dispatched ceremonially is
apparent in the field notes. Referring to burial 39-6, engineer and mapper Eduardo
Contreras comments, "Del entierro no. 39, solamente se encontro una parte que fue:
la calavera, parte de la columna vertebral y las costillas no apareciendo el resto del
esqueleto. Seguramente fue mutilado" [from burial no. 39, only a portion was found:
the skull, part of the spinal column and ribs; the rest of the skeleton was not apparent.
Certainly it was mutilated] (my translation) (Contreras n.d.:VII:5l). Referring to
burial 12-6 (IS), Contreras states: "El craneo, como la mayoria de los huesos, estan en
mal estado de conservacion y no se encontraron los huesos de las extremidades
inferiores solamente las apofisis de los femurs (hip joints are illustrated). Da la
impresion de haber sido mutilado antes de ser enterrado" [the skull, like the majority
of the bones, is in a poor state of preservation and the bones of the lower extremities
were not found, only the joints of the femurs. It gives the impression of having been
mutilated before burial] (my translation) (Contreras n.d.:VII:37). Certainly the burial
of some people under a ballcourt and around a stone pillar footing are unique among
the Casas Grandes burials. The disarticulation and absence of portions of the skeleton
as noted by Contreras above, and of one of the females under the ballcourt along with
the burial context make a compelling argument for human sacrifice, but the lack of
ritual paraphernalia in these burials renders this interpretation speculative.
Cannibalism in archaeological bone was not widely regarded until after the
publication of the Casas Grandes site report. The NIS assemblage was never studied
96
in earnest during preparation of the site report or by subsequent researchers. Thus, it
is not surprising that this type of damage went unreported. Comparison of
postmortem modification at Paquime to that found on non-cannibalized assemblages
reveals several disparities, while comparison to Southwest cannibalism sites in Table
7, above, reveals strong similarities. The assemblage from La Quemada (Nelson et al.
1992) shows no evidence of disarticulation, fragmentation, scalping or burning, and
the skeletons were relatively complete. This is more typical of secondary burial. The
massacre victims from Crow Creek (Willey 1990) and Larson (Owsley et al. 1977)
showed evidence of mutilation (dismemberment, decapitation, anterior tooth avulsion,
nose and tongue removal, and scalping), but lacked evidence of the extensive bone
fragmentation seen at Paquime. Also note that Crow Creek and Larson showed a
good deal of carnivore gnawing to the ends of the longbones, which Paquime lacked.
The Ram Mesa assemblage (Ogilvie and Hilton 2000) interpreted as the remains of
murdered witches, showed cranial trauma and cutmarks, anterior tooth avulsion, and a
high incidence of burning.
Therefore, the remains from Paquime most resemble those from Mancos and
other sites with remnants of cannibalism (refer to Table 7). It is noteworthy that the
time period during which this activity occurred (the Medio period) coincides with an
increase in violence and cannibalism in the Four Comers of the United States
Southwest.
In summary, Hypotheses 1 and 2 were not supported by these data. The IS
assemblage shows only one conclusive case of cranial trauma, which could also be
97
explained by intragroup hostility. Furthermore, the low incidence of antemortem
postcranial fractures, which can be explained as accidental, is not different from other
sites lacking warfare. Massacre is also unlikely as an explanation for the NIS
assemblage. The low incidence of distinguishable trauma from the NIS and the
similarity of trauma occurrence between the IS and NIS argue against a massacre.
The fluorine and AMS dates are also inconsistent with a massacre at the terminus of
the city's occupation. Hypothesis 3 was validated by the evidence of extreme
postmortem modification that is similar to sites where cmmibalism occurred. The
possibility of ritual sacrifice still remains a viable hypothesis due to the burial
provenience of several bodies as DiPeso noted.
98
CHAPTER VII
DISCUSSION
Warfare
Both warfare and massacre in the Casas Grandes valley are predicated on
DiPeso's belief that Toltec pochteca were controlling the city and subjugating the
region to its power. DiPeso thought that violence was used to subdue the regional
polities. Since the dating sequence of the site was revised in 1993, this model was
invalidated and an alternative model to explain the motive for this violence has not
been offered. Though some Mesoamerican cultural elements were incorporated by
Paquime's elite, it does not follow that the city was under the control of a
Mesoamerican state. Perhaps there were periodic skirmishes in the area, but this
study found no evidence of ongoing warfare.
As mentioned in Chapter 1, LeBlanc (1999:200-203) maintains that Late
Period settlement patterns and locations changed in the Southwest due to increased
hostility. With the exception of the large apartment cluster, the layoutofPaquime
does not fit this model for four reasons: 1) units 1, 11, 15, 18,20, 21, and 23 are free-
standing (refer to Figure 4 above), 2) small Medio period sites are scattered
throughout the Casas Grandes and other nearby valleys, 3) the ballcourts and plazas
are not enclosed within any walled structures, and 4) Paquime is located in the middle
of a flat floodplain, implying a choice of proximity to agriculture rather than to safety.
These attributes make Paquime a poor candidate for a defensively structured
99
settlement. However, it could be argued that the hilltop communication station of
Cerro de Moctezurna mitigated any offensive threat by providing advance warning of
approaching enemies.
The low number of burials and disparate sex ratio at Paquime are unusual but
not unique. Swarts Ruin, a Mimbres site in southwest New Mexico was characterized
by a similar burial pattern and low number ofobserved burials (Cosgrove and
Cosgrove 1974). The Cosgrove's (1974:24) note that some rooms contained no
burials, while others had between 1 and 32. Eighteen and one half percent of the Late
period burials were from outside structural walls or in room fill, which compares to
22% NIS from Paquime. Nineteen of these bodies were from the fill of rooms that
they state "probably had fallen into disuse as the population decreased before the final
abandonment of the village... since it was probable that numbers of houses fell into
disuse and were left unoccupied during the time necessary to build settlements of the
Early and Middle Periods" (Cosgrove and Cosgrove 1974:24). Also, the site had
been extensively burned..
The Point of Pines site in southeast New Mexico shows a similar situation
(Bennett 1973). The Late period burials (A.D. 1285-1450) from Point of Pines
showed a bias of 36% females to 30% males, or 84 males to 100 females, which is
comparable to Paquime's 65 males to 100 females. Bennett (1973:53) found no
significant ditTerence between these sex ratios using the chi-square test.
Some possibilities to explain the lack of burials are: 1) some people
responsible for production in the city lived in other regional hamlets, 2) many of the
100
men traveled as traders rather than warriors and died while away, 3) some people are
buried outside the excavation area as seen at Swarts Ruin, 4) the site experienced
population fluctuation throughout the year, which may have been tied to ceremonial
and/or trade gatherings, and 5) the population declined over time and all rooms were
not used throughout the entire occupation of the city. Additionally, the NIS should be
included in the total count of Medio period burials and not treated as a separate
population, which would increase the number of burials.
Massacre
The absence of warfare makes justifying a massacre difficult. Attacking the
city in the absence of hostility defies logic. If there were no pochteca subduing the
region's people, what purpose would invading the richest city in one of the most
productive regions in the area be if you abandoned it and a wealth of commodities
immediately afterwards? The risk certainly outweighs the reward.
This leaves the difficult task of explaining the differences among and between
the IS and NIS groups. DiPeso originally subsumed all NIS (with the exception of
three individuals) into the homogenous type 2 burial group. These remains further
lost their uniqueness when he assigned all of them to the Diablo phase of the Medio
period. In fact, one individual (18-16) was buried so high in room infill that it led
DiPeso to comment "it is hard to believe that it belongs with any occupation of the
site (except perhaps the modem Mexican occupation)" (DiPeso n.d.: XII: 155). In the
site report this individual is listed as associated with the Diablo phase ofthe Medio
period (DiPeso et al. 1973:8:404). Deconstructing the type 2 group may add clarity.
101
I consider the articulated NIS (types 2A, 2C) to have been formal interments
post-dating the occupation ofthe rooms in which they were buried and possibly post-
dating the occupation of the city. Uncannibalized individual 26-14 (type 2A) was
found with his arms folded across his chest; cannibalized individual 27-14 (type 2C)
was found associated with the remains ofthree turkeys, implying formalized burial.
It appears that the fragmentary, disarticulated remains from floor surfaces and mixed
in fill above the floors (types 2B, 2D) were either deposited during the occupation of
the city, but after the rooms in which they were buried fell into disuse, or after the
abandonment of the city. Using their depositional provenience to bolster the
massacre hypothesis makes little sense due to their occurrence in room fill above the
floor surfaces (in some cases as high as 40 cm and with ceiling tiles found in the fill
below them). Interpretation ofthe single type 2E burial is difficult due to its recovery
from a test trench.
Thus, burial context of the NIS appears consistent with the occupants' use of
the crumbling and abandoned rooms and plazas as burial sites. Rather than
representing a single depositional episode, the burial data from the NIS favor
continuous deposition during and/or after site occupation. The frequency ofNIS
deposited above floor surfaces in fill (57%) is equal to the amount ofNIS
cannibalized. The only NIS with evidence of cannibalism were fragmentary,
disarticulated skeletons.
Burial types 2A and 2C were simply individuals who were buried in filled-in
abandoned houses, where their pits were either dug partially through the soft: fill or
102
until reaching the hard floor surface. They may have all been dug to the floor surface,
with the pits of the 2C burials filling partially with back dirt between the time of pit
preparation and actual deposition of the body. Therefore, the reason for the
abandonment of the city remains unknown.
Human Sacrifice and Cannibalism
Making the leap from a highly integrated ritual system to a system that
included human sacrifice and ceremonial cannibalism is difficult. Ritual human
sacrifice in Mexico is well known in the archaeological literature and in central
Mexico it was often tied to the ballgame. Human sacrifices have been found in
Mexico in both apartment compounds and monuments at Teotihuacan (Sugiyama
2005), and at Classic Maya sites (Tiesler and Cucina 2006). Tenth-century sculptures
at El Tajin on the east coast of Mexico depict sacrifices by decapitation and
disembowelment at the conclusion of ballgames (Wilkerson 1991 :63). DiPeso et al.
(1974:8) describe ten Paquime burials that may have been human sacrifices. The
burial of a select few under the ceremonial ballcourt makes a compelling argument
for ritual human sacrifice tied to the ballgame. It is also possible that some of the
cannibalized group were victims of ritual and/or ballgame-related sacrifice.
Like others in the Southwest and northern Mexico, the people of Paquime
were primarily dependent on agriculture for their existence. This period was
characterized by unpredictable climatic fluctuation, which likely caused stress on the
population. Control over natural events is a primary reason why people seek
supernatural assistance (Bailey and Peoples 1999:211). Human sacrifices during
103
ballgames were often tied to ensuring fertility or adequate precipitation (Wilkerson
1991). At El Tajin, the ballgame and the sacrifice that followed were necessary steps
to send mortals to the underworld to confer with the deities (Wilkerson 1991 :65).
The possibility that the ballgame played a role in perpetuating the sacrifices at
Paquime cannot be ruled out.
My data show that cannibalism was practiced at Paquime since at least the
beginning of the Media period. Interpretation is complicated due to recovery of
cannibalized remains from both interred and non-interred contexts. One can only
speculate as to the reasons why the people of Paquime practiced conspecific
consumption. I will offer some possibilities.
The evidence suggests there may have been more than one type of
cannibalism practiced: endocannibalism, and exocannibalism, ceremonial and profane
cannibalism, or a combination of both. Goldman (1999: 14) defines endocannibalism
as the consumption of a member of one's own group (through kinship or descent);
exocannibalism refers to consumption of people from outside one's own group
(enemies, for example). Profane or survival cannibalism is the consumption of
human flesh in emergency starvation situations, whereas any consumption of human
flesh during a ritual or for religious reasons is labeled ceremoniaL The IS
cannibalized skeletons may have been the remains of endocannibalism, or ceremonial
cannibalism (or endo-ceremonial cmmibalism), while the cannibalized NIS are those
of outsiders or those consumed profanely for nutrition. Most consumed IS were more
complete than their NIS counterparts, which suggests the remains of at least some
104
individuals were important enough to warrant their collection and burial in a single
sub-floor grave, which requires more energy expenditure (Ravesloot 1988). This
courtesy may have been afforded to certain lineages or group members, or those who
were sacrificed ceremonially. Burial 44 A-L-13 serves as an example.
Burial 44-13 was a sub-floor (type 4E) pit sealed with wood planks (Figure
27). The non-permanent sealing with planks implies that it was re-opened to deposit
bodies through time. It contained at least 12 individuals (most of them articulated)
and the greatest number and quality of artifacts of any burial at Paquime (DiPeso et al.
1974:8:387-388). The pit had a shelf built into one end and ceremonial ceramic
drums among the artifacts. The articulated individuals in this pit were piled on top of
one another with the uppermost layer containing the fragmentary remains of four
adolescents. The layering of bodies and artifacts was so impressive it led DiPeso to
state it was "the most complex burial discovered at Casas Grandes" (DiPeso et al.
1974:8:387). Though I did not personally examine any ofthe 44-13 material, Sophie
Kohn of the University of New Mexico (personal communication 2008) examined the
remains from the pit and forwarded digital images to me. We both believe the
fragmentary material from the uppermost level of this burial exhibits a similar pattern
of cutmarks, burning, and longbone fragmentation to those found at cannibalism sites.
This is due to similarities in processing damage between the 44-13 material and the
36 skeletons with PM reported here. I believe this burial is good evidence of the
burial of an elite lineage coupled with endo- or ceremonial cannibalism.
105
CEl2AJI\lC IPCJ"ITlFICA.TION
o
r-CG/.r(:3.o9
2 - CG/430a
3-CG/4301
4- CG/4300
S-CG/4307
6 - CG/430Z
7- CG/430S
a-CG/4306
9 -CG/4304
Figure 27. Complex burial 44 A-L-13 (DiPeso et al. 1974:8:Figure 383-8).
106
The cannibalized NIS were denied sub-floor burial, were less complete, and
scattered, and discarded on floors or in fill of unused buildings. This disregard for
their final disposition shows the relative unimportance ofthe persons consumed,
inferring their status as an outsider or the consumption as profane in nature. Burial
27-14 is enigmatic in this respect due to its relative incompleteness and placement on
a floor surface in association with the remains of three turkeys (a sacred animal
associated with death). However, data collection from this skeleton was minimal
since only five percent of this individual was recovered.
Element Representation ofFragmented Skeletons
Turner and Turner (1999) cite under-representation of ribs and vertebrae as a
characteristic of cannibalized bone assemblages. The Paquime assemblage did show
under-representation of vertebrae, but the observed frequency of ribs was nearly
identical to the expected frequency. It is understood that this method of comparison
is somewhat hypothetical in that it compares frequencies of highly fragmented
archaeological bone to those occurring in an undamaged, complete skeleton.
However, it is curious that the two most over-represented categories (cranial and
longbone) were the most fragmented, and the most under-represented category (hands
and feet) contains some ofthe smallest bones, and hence the most difficult to recover
archaeo10gically.
The comparison ofskeletal element representation showed no significant
difference between the expected and observed frequencies of different skeletal
elements. The largest pair-wise differences were in the overabundance of cranial and
107
10ngbones, and the lack of vertebrae, and hand and foot bones. These differences can
be explained by the fact that the crania and longbones were the most highly
fragmented parts of the skeleton and that the bones of the hands and feet are the most
difficult to recover due to their diminutive size and their susceptibility to the
processes of diagenesis. Note that in Table 7, all but one site lists under-
representation ofvertebrae, and hands and feet as an attribute of cannibalism. This
implies that the phenomenon is cornmon and whether or not it is due to loss during
the postmortem body preparation or as the result of diagenesis and/or archaeological
recovery techniques is difficult to discern. This exercise does show that the crania
and longbones were more highly fragmented than expected.
Non-Osteological Evidence: Chronometric Dating and Burned Structures
Relative and absolute dating methods often yield the best information for
reconstructing a timeline of past activities. The fluoride and AMS data have been
enlightening in the case of Paquime. It was expected that fluoride ion concentrations
between the IS and NIS would be different, when in fact the two were quite similar.
The NIS showed no horizontal tendency that would indicate a single depositional
episode as would be expected in the case of a massacre.
AMS dating was applied to NIS samples to indicate an absolute date for the
proposed massacre. The resulting dates corresponded to the Medio period as
expected, and did show a considerable degree of overlap. However, I would feel
more comfortable making a massacre assessment with additional dates. A better
pattern of deposition may emerge with analysis ofmore samples.
108
Burning of structures during prehistory in the Southwest was not unusual.
Ritual closure by burning was commonly practiced on ceremonial, communal, and
housing structures. According to LeBlanc (1999:74):
"There should be only a few explanations for a site or a portion of a
site to have burned. In general, the logical possibilities for burning
sites or portions of sites are (1) accidental fires; (2) deliberate burning
of sites that was not related to warfare; (3) deliberate burning of
particular structures (such as kivas) while the community continued to
be occupied-again, unrelated to warfare-(reasons for this burning
might include ceremonial purposes, internal community violence,
death of its inhabitants, or pest removal); and (4) burning of all or part
of the community due to warfare-including the special case of
burning by the resident group to deny the site to the enemy."
However, no one has attempted to explain repeated burning of structures, a
phenomenon that was recorded in some rooms at Paquime.
DiPeso (1974:2:325) argues that the burned room blocks at Paquime are
evidence of the massacre that ended the site's occupation. In his view, invaders
stormed the city, murdered the final population, desecrated sacred objects, and burned
the buildings. Yet, in his excavation notes on file in the Amerind Foundation
archives, he notes that some rooms were burned more than once and many had
multiple floor surfaces superimposed on each other. Rooms 4 and 17 of unit 16 were
noted to have been burned, re-plastered, then burned a second time (DiPeso
n.d.:XII:57, 114). Room 3 in unit 16 contains three superimposed floor surfaces and
was also burned, then re-occupied (DiPeso n.d.:XII:45). None ofthe 8 individuals
from unit 16 that I viewed (IS or NIS) showed any evidence of burning, suggesting
that the rooms burned before the bodies were deposited there. I conclude that the
109
fires were more likely tied to ritual closures rather than a site-wide burning that would
indicate warfare, massacre, or denial of the site to the enemy.
110
CHAPTER VIII
CONCLUSION
The objectives of this dissertation were to determine whether inter-group
violence was endemic in the Casas Grandes region, whether the final population was
massacred, and whether the people of Paquime practiced human sacrifice and
cannibalism. These questions are tied to the more general objective of explaining the
cultural dynamics surrounding violence in Northwest Mexico and why the city was
abandoned. Multiple lines of inquiry were used to test these hypotheses. Three
conclusions were reached. First, the frequency of skeletal injury characterizes a non-
warring population. The antemortem fractures can be more parsimoniously explained
as accidental. Second, the osteological and burial data, chronometric dates, and
evidence of structural buming do not support the massacre hypothesis. Rather, the
data support the following conclusion: the articulated and disarticulated, non-interred
skeletons are primary and <secondary burials, respectively, made in the soft fill of
abandoned structures during and after the occupation of the city. The multiple
periods of burning are more characteristic of ritual room closure rather than a single
massacre episode.
Finally, very strong evidence for ritual sacrifice and cannibalism anlong both
the non-interred and formally interred assemblages was found. Given the large
amount of symbolically-charged Casas Grandes polychrome pottery found at
111
Paquime and other regional sites, ritual-religious ceremony was likely a large part of
daily life. The strong connections between Mesoamerican ballcourts and ritual are
also well-documented. At a minimum, the burials under the ballcourt and the child
burial around the post in room 21 C of unit 8 should be considered to have been
ceremonial in nature. The difference in treatment of certain cannibalized skeletons
(the more complete IS) should also be considered to have been surrounded by
ceremony. These conclusions should create renewed interest in examination of
various aspects of Casas Grandes culture, allowing for a greater understanding of the
people who occupied this site for 250 years. The data only discredit the theories of
warfare and massacre, but do not offer an explanation for the abandonment of the site.
Further research may answer that question.
Contributions ofThis Research
Though recovered over 40 years ago, this is the first detailed examination of
the non-interred Medio period skeletal material. This analysis serves as an example
that even fragmentary skeletal remains can be an important data source and should
not be neglected in favor of more complete remains with better provenience. This
dissertation has contributed 1) an analysis of the non-interred skeletal remains, and
their burial provenience, 2) recognition of cannibalism at Paquime, 3) new fluoride
ion and AMS radiocarbon dates from human skeletal remains, and 4) analysis of
skeletal element representation in cannibalized assemblages. It has also raised two
new questions: 1) is the female sex bias real or introduced, and 2) did Paquime
function as more of a regional gathering place and ceremonial center rather than
112
simply a manufacturing and distribution center? Certainly, there is much work left to
do at Paquime.
Future Research Directions
The fact that the NIS were highly fragmented and incomplete, suggests the
remains may represent fewer individuals that were scattered after disarticulation. The
only means of demonstrating this would be to re-assess the number of individuals by
conjoining the fragments of the NIS assemblage. I suspect this would produce fewer
individuals.
APPENDIX A
INDIVIDUALS INCLUDED IN STUDY SAMPLE
113
Burial Burial Percent Burial Burial Percent
Number Unit Type Complete Number Unit Type Complete
2 1 1B 25 13 6 1A 25
3 1 1B 25 15 6 1B 50
4 1 1B 25 17 6 1B 20
9 1 1A 20 20 6 1B 50
12 1 1A 30 21 6 1A 90
13 1 1B 10 22 6 1B 90
14 1 2 5 24 6 2 <5
15 1 2 20 27 6 1B 15
16 1 1A 10 28 6 1B 20
20 1 2 <5 29 6 1B 20
23 1 3A 50 31 6 1B 40
28 1 3A <5 35 6 3A 50
30 1 3D <5 40 6 3H 15
17A 1 1B 20 43 6 1A 40
17B 1 1B 15 47 6 1A 75
19A 1 2 10 48 6 1A 50
19B 1 2 10 49 6 3B <5
19C 1 2 <5 23A 6 1B 10
1A 1 2 10 23B 6 1B <5
21A 1 2 15 23C 6 1B 10
30C 1 3D <5 37A 6 3A 35
30D 1 3D <5 37B 6 3A 5
30E 1 3D <5 5A 6 2 15
30G 1 3D <5 5B 6 2 <5
1 4 2 80 5 8 1A 35
4 6 2 <5 20 8 3A 10
6 6 3A 50 23 8 8A 90
8 6 1A 25 24 8 1A 90
12 6 1H 5 25 8 1A 50
114
Burial Burial Percent Burial Burial Percent
Number Unit Type Complete Number Unit Type Complete
27 8 1A 60 25 12 12A 25
28 8 2 15 31 12 2 10
29 8 2 <5 34 12 11B 60
31 8 2 10 29A 12 2 25
33 8 1A 10 6A 12 1B 15
44 8 'IB 25 6B 12 1B 25
45 8 1B 5 1 13 2 15
26A 8 1A 40 4 13 2 5
26B 8 1A 5 5 13 2 5
1 11 2 15 12 13 1B 50
2 11 1A 40 13 13 4A 40
3 11 1E 10 14 13 11D 10
8 11 2 10 15 13 11 D 10
9 11 2 20 16 13 11D <5
32 11 1B 40 19 13 3A 15
33 11 1B 30 23 13 1B 25
35 11 7A 25 24 13 1A 60
36 11 'IB 25 25 13 3A 20
37 11 1A 30 26 13 1B 50
43 11 1A 20 45 13 11B <5
45 11 7A 10 46 13 1'I B 10
52 11 9 <5 72 13 3B 50
54 11 7A 25 73 13 3B 35
18/19 11 2 <5 74 13 3A 50
53A 11 7B 90 76 13 1A 35
4 12 1A <5 77 13 1A 30
4 12 1A 30 78 13 3A 35
15 12 1A <5 81 13 3A 35
16 12 1A <5 82 13 3A 25
24 12 12A 20 86 13 3A 50
115
r-:-urial Burial Percent Burial Burial Percent
Number Unit Type Complete Number Unit Type Complete
87 13 3A 40 98 15 10 10
88 13 3A 40 1 16 2 50
488 13 118 25 8 16 9 10
62A 13 18 15 14 16 2 5
66A 13 1G 30 18 16 2 <5
f------
668 13 1G 40 20 16 1A 30
1 14 88 40 21 16 1A 60
2 14 88 50 23 16 48 40
--
19 14 2 10 25 16 1A 15
23 14 2 <5 3 19 1E <5
25 14 2 5 6 19 11A 25
26 14 2 25 7 19 1A 20
27 14 2 5 8 19 2 <5
28 14 2 <5 1 20 1H 40
31 14 18 40 2 20 1A 25
32 14 18 40 1 21 1A 10
39 14 78 40 4 21 38 <5
44 14 18 5 1 22 3A 30
45 14 18 10 2 22 3A 20
46 14 18 50 1 23 38 <5
48 14 10 50 4 23 3A 15
49 14 10 75 3 CP 38 20
51 14 12A 30 4 CP 38 10
53 14 11A 25 5 CP 38 <5
54 14 11A 20 11 CP 3A 50
--
2 15 2 30 21A CP 3D <5
3 15 2 5 218 CP 3D <5
3 15 2 5 210 CP 3D <5
10-900874
4 15 2 5 0/4 TROPHY SKULL <5
5 15 2 <5
9A 15 10 20
APPENDIXB
DETAILS OF FLUORINE ANALYSIS
Average Coefficient
% % Standard of
Sample ID Fluoride Fluoride Deviation Variation
19C-1 0.730
19C-1 0.797
19C-1 0.828 0.785 0.050 6.4
24-6 0.557
24-6 0.566
24-6 0.569 0.564 0.006 1.1
18/19-11 0.460
18/19-11 0.510
18/19-11 0.487 0.486 0.025 5.2
21D-CP 0.870
21D-CP 0.864
21D-CP 0.850 0.861 0.010 1.2
21B-CP 0.774
21B-CP 0.733
21B-CP 0.696 0.735 0.039 5.3
25-14 0.949
25-14 0.960
25-14 0.940 0.950 0.010 1.1
27-14 1.060
27-14 1.065
27-14 1.101 1.076 0.023 2.1
30-1 0.336
30-1 0.416
30-1 0.288 0.347 0.065 18.6
11-8 1.036
11-8 1.049
11-8 1.020 1.035 0.014 1.4
44-8 0.584
116
117
44-8 0.557
44-8 0.560 0.567 0.015 2.6
19-13 0.962
19-13 0.917
19-13 0.966 0.949 0.027 2.9
19A-1 0.266
19A-1 0.278
19A-1 0.221 0.255 0.030 11.8
21-1 0.375
21-1 0.387
21-1 0.381 0.381 0.006 1.5
9A-15 0.410
9A-15 0.378
9A-15 0.353 0.380 0.029 7.5
1-11 0.482
1-11 0.514
1-11 0.505 0.501 0.017 3.3
31-12 0.740
31-12 0.726
31-12 0.832 0.766 0.058 7.5
17-6 0.225
17-6 0.194
17-6 0.176 0.198 0.025 12.7
5-6 0.385
5-6 0.371
5-6 0.293 0.350 0.050 14.2
30-1 0.337
30-1 0.276
30-1 0.325 0.313 0.033 10.4
118
BIBLIOGRAPHY
Aitken, Martin J.
1990 Science-based Dating in Archaeology. Longman, London.
Anyon, Roger, and Steven A. LeBlanc
1984 The Galaz Ruin: A Prehistoric Mimbres Village in Southwestern New
Mexico. Maxwell Museum ofAnthropology Publication Series. Maxwell
Museum of Anthropology and the University ofJ"Jew Mexico, Albuquerque.
Arens, William
1993 The Man-Eating Myth: Anthropology and Anthropophagy. Oxford
University, New York.
Bailey, Garrick and James Peoples
1999 Introduction to Cultural Anthropology. West/Wadsworth, Belmont,
California.
Baker, Brenda J., Tosha L. Dupras, and Matthew W. Tocheri
2005 The Osteology of Infants and Children. Texas A&M Anthropology Series,
edited by D. Gentry Steele. Texas A&M University, College Station.
Bandelier, Adolph F.
1890 Final Report of Investigations Among the Indians of the Southwestern
United States, Carried on Mainly in the Years 1880-1885. Part 1.
Archaeological Institute of America Papers, American Series. Cambridge
University Press, Cambridge, Massachusetts.
Barker, Francis, Peter Hulme, and Margaret Iversen
1998 Cannibalism and the Colonial World. Cambridge University, Cambridge,
U.K.
Benfer, Robert A.
1968 An Analysis ofa Prehistoric Skeletal Population, Casas Grandes,
Chihuahua, Mexico. Ph.D. dissertation, Department of Anthropology,
University of Texas, Austin.
Bennett, Kenneth A.
1973 The Indians ofPoint ofPines, Arizona. University of Arizona, Tucson.
119
Benson, Elizabeth P., and Elizabeth H. Boone
1984 Ritual Human Sacrifice in Mesoamerica. Dumbarton Oaks, Washington,
D.C.
Berryman, Hugh E. and Steven A. Symes
1998 Recognizing Gunshot and Blunt Cranial Trauma Through Fracture
Interpretation. In Forensic Osteology: Advances in the Identification of
Human Remains, edited by Kathleen J. Reichs, pp. 333-352. 2nd ed. Charles.
C. Thomas, Springfield, Illinois.
Billman Brian R., Patricia M. Lambert, and Banks L. Leonard
2000 Cannibalism, Warfare, and Drought in the Mesa Verde Region During the
Twelfth Century A.D. American Antiquity 65(1): 145-178.
Bocquet-Appel, Jean-Pierre, and Claude Masset
1982 Farewell to Paleodemography. Journal ofHuman Evolution 11 :321-333.
Brand, Donald D.
1935 Distribution of Pottery Types in Northwest Mexico. American
Anthropologist 37:287-395.
Bronk Ramsey, Christopher
2001 Development of the Radiocarbon Calibration Program OxCal. Radiocarbon
43:355-363.
Bronk Ramsey, Christopher
1995 Radiocarbon Calibration and Analysis of Stratigraphy: The OxCal Program.
Radiocarbon 37:425-430.
Brooks, S., and Judy M. Suchey
1990 Skeletal Age Determination Based on the Os Pubis: A Comparison of the
Acsadi-Nemeskeri and Suchey-Brooks Methods. Human Evolution 5:227-238.
Buikstra, Jane E. and Douglas H. Ubelaker
1994 Standards for Data Collection from Human Skeletal Remains. Arkansas
Archeological Survey Research Series, edited by Hester A. Davis. Arkansas
Archeological Survey, Fayetteville.
Buikstra, Jane E. and Lyle W. Konigsberg
1985 Paleodemography: Critiques and Controversies. American Anthropologist
87:316-333.
r .
120
Butler, Barbara H.
1971 The People ofCasas Grandes: Cranial and Dental Morphology Through
Time. Ph.D. dissertation, Southern Methodist University, University Park,
Texas. University Microfilms, Ann Arbor.
Caceres, Isabel, Marina Lozano, and Palmira Saladie
2007 Evidence for Bronze Age Cannibalism in El Mirador Cave (Sierra de
Atapuerca, Burgos, Spain). American Journal ofPhysical Anthropology
133:899-917.
Carey, H. A.
1931 An Analysis of the Northwestern Chihuahua Culture. American
Anthropologist 33(3):325-374.
Carreta, M. Nicolas
2008 Archaeology and Physical Anthropology: A Reflection on Warfare in the
Archaeological Vision. In Archaeology Without Borders: Contact, Commerce,
and Change in the Us. Southwest and Northwestern Mexico, edited by Laurie
D. Webster, Maxine E. McBrinn, and Eduardo Gamboa Carrera, pp. 385-392.
University of Colorado, Boulder.
Clark, Steven C., Mary Fran Ernst, William D. Haglund, and Jeffrey M. Jentzen
1996 The Medicolegal Death Investigator: A Systematic Training Program for
the Professional Death Investigator. Occupational Research and Assessment,
Inc., Big Rapids, Michigan.
Coe, Michael D.
1994 Mexico: From the Olmecs to the Aztecs. 4th ed. Thames and Hudson, New
York.
Coe, Michael D., and Rex Koontz
2002 Mexico: From the Olmecs to the Aztecs. 5th ed. Thames and Hudson, New
York.
Contreras, Eduardo
n.d. Casas Grandes Excavation Fieldbook VII. Amerind Foundation archives,
Dragoon, Arizona.
Cordell, Linda S.
1997 Archaeology ofthe Southwest. 2nd ed. Academic Press, San Diego,
California.
Cordell, Linda S.
1984 Prehistory ofthe Southwest. Academic Press, Orlando.
121
Corruccini, Robert S.
1983 Pathologies Relative to Subsistence and Settlement at Casas Grandes.
American Antiquity 48:609-610.
Cosgrove, Harriet S. and Charles B. Cosgrove
1974 The Swarts Ruin: A Typical Mimbres Site in Southwestern New Mexico.
No.1. Papers of the Peabody Museum of American Archaeology and
Ethnology, Harvard University. vol. XV, Kraus Reprint, New York.
Dales, George F.
1979 The Mythical Massacre at Mohenjo-Daro. In Ancient Cities ojthe Indus,
edited by Gregory L. Possehl, pp. 293-296. Carolina Press, Durham, North
Carolina.
Darling, J. Andrew
1998 Mass Inhumation and the Execution of Witches in the American Southwest.
American Anthropologist 100:732-752.
Davis-Salazar, Karla L.
2007 Ritual Consumption and the Origins of Social Inequality in Early
Formative Copan, Honduras. In Mesoamerican Ritual Economy:
Archaeological and Ethnological Perspectives, edited by E. Christian Wells
and Karla L. Davis-Salazar, pp. 197-220. University of Colorado, Boulder.
Dean, Jeffrey S.
1996 Demography, Environment, and Subsistence Stress. In Evolving Complexity
and Environmental Risk in the Prehistoric Southwest, Proceedings Volume
XXIV, Santa Fe Institute Studies in the Sciences ojComplexity, edited by
Joseph A. Tainter and Bonnie Bagley Tainter, pp. 25-56. Addison-Wesley,
Reading, Pennsylvania.
Dean, Jeffrey S. and John C. Ravesloot
1993 The Chronology of Cultural Interaction in the Gran Chichimeca. In Culture
and Contact: Charles C. DiPeso's Gran Chichimeca, edited by Anne 1.
Woosley and John C. Ravesloot, pp. 83-103. Amerind Foundation, Dragoon,
Arizona.
DiPeso, Charles C.
1974 Casas Grandes: A Fallen Trading Center of the Gran Chichimeca. vol. 1-3,
Amerind Foundation, Dragoon, Arizona.
DiPeso, Charles C.
n.d. Casas Grandes Excavation Fieldbook XII. Amerind Foundation archives,
Dragoon, Arizona.
122
DiPeso, Charles C., John B. Rinaldo, and Gloria J. Fenner
1974 Casas Grandes: A Fallen Trading Center of the Gran Chichimeca. vol. 4-8,
Amerind Foundation, Dragoon, Arizona.
Douglas, John E.
1995 Autonomy and Regional Systems in the Late Prehistoric Southern
Southwest. American Antiquity 60:240-257.
Dunbar, Robin 1. M.
1984 Reproductive Decisions. Princeton University, Princeton.
Edgar, Heather J. H., and Paul W. Sciulli
2006 Comparative Human and Deer (Odocoileus virginianus) Taphonomy at the
Richards Site, Ohio. International Journal ofOsteoarchaeology 16:124-137.
Ekholm, G. F.
1940 The Archaeology of Northern and Western Mexico. In The Maya and Their
Neighbors, edited by Hays, C., pp. 320-330. D. Appleton-Century, New York.
Fash, Barbara W., and Willaim L. Fash
2007 The Roles ofBallgames in Mesoamerican Ritual Economy. In
MesoamericanRitual Economy: Archaeological and Ethnological
Perspectives, edited by E. Christian Wells and Karla L. Davis-Salazar, pp.
267-286. University of Colorado, Boulder.
Flinn, Lynn, Christy G. Turner II, and Alan Brew
1976 Additional Evidence for Cannibalism in the Southwest: The Case of LA
4528. American Antiquity 41(3):308-318.
France, Diane L.
1998 Observational and Metric Analysis of Sex in the Skeleton. In Forensic
Osteology: Advances in the Identification ofHuman Remains, edited by
Kathleen J. Reichs, pp. 163-186. 2nd ed. Charles C. Thomas, Springfield,
Illinois.
Frayer, David W.
1997 Ofnet: A Mesolithic Massacre. In Troubled Times: Violence and Warfare in
the Past, edited by Debra L. Martin and David W. Frayer, pp. 181-216.
Gordon and Breach, Amsterdam.
123
Gaither, Catherine, Jonathan Kent, Victor Vasquez Sanchez, and Teresa Rosales
Tham
2008 Mortuary Practices and Human Sacrifice in the Middle Chao Valley of Peru:
Their Interpretation in the Context of Andean Mortuary Patterning. Latin
American Antiquity 19(2): 107-122.
Goldman, Laurence R.
1999 From Pot to Polemic: The Uses and Abuses of Cannibalism. In The
Anthropology ofCannibalism, edited by Laurence R. Goldman, pp. 1-26.
Bergin and Garvey, Westport, Connecticut.
Haddy, A. and A. Hanson
1982 Nitrogen and Fluorine Dating of Moundville Skeletal Samples.
Archaeometry 24:37-44.
Hammond, George P., and Agapito Rey
1928 Obregon's History of16th Century Explorations in Western America,
Entitled Chronicle, Commentary, or Relation ofthe Ancient and Modern
Discoveries in New Spain and New Mexico, Mexico, 1584. Wetzel, Los
Angeles.
Haverkort, Caroline M., and David Lubell
1999 Cutmarks on Capsian Human Remains: Implications for Haghreb Holocene
Social Organization and Palaeoeconomy.International Journal of
Osteoarchaeology 9:147-169.
Hurlbut, Sharon A.
2000 The Taphonomy of Cannibalism: A Review of Anthropogenic Bone
Modification in the American Southwest. International Journal of
Osteoarchaeology 10:4-26.
Judd, Margaret A. and Charlotte A. Roberts
1999 Fracture Trauma in a Medieval British Farming Village. American Journal
ofPhysical Anthropology 109:229-243.
Jurmain, Robert
1991 Paleoepidemiology of a Central California Prehistoric Population from CA-
A1a-329. In Paleopathology: Current Synthesis and Future Options, edited by
Donald J. Ortner and A.C. Aufderhiede, pp. 241-248. Smithsonian Institution,
Washington, D.C.
Jurmain, Robert
1999 Stories From the Skeleton: Behavioral Reconstruction in Human Osteology.
Gordon and Breach, Amsterdam.
124
Kelley, J. Charles
2000 The Aztatlan Mercantile System: Mobile Traders and the Northwestward
Expandion of Mesoamerican Civilization. In Greater Mesoamerica: The
Archaeology ofWest and Northwest Mexico, edited by Michael S. Foster and
Shirley Gorenstein, pp. 137-154. University of Utah, Salt Lake City.
Kennedy, Kenneth A.R.
1984 Trauma and Disease in the Ancient Harrapans. In Frontiers in the Indus
Civilization, edited by B.B. Lal and S.P. Gupta, pp. 425-436. Books and
Books, New Delhi.
Kidder, Alfred V.
1916 The Pottery of the Casas Grandes District, Chihuahua. Peabody Museum of
Archaeology and Ethnology. Harvard University Press, Cambridge,
Massachusetts.
Konigsberg, Lyle W, and Samantha M. Hens
1998 Use of Orginal Categorical Variables in Skeletal Assessment of Sex from
the Cranium. American Journal ofPhysical Anthropology 107:97-112.
Kroeber, Alfred L.
1928 Native Culture ofthe Southwest. University of California, Berkeley.
Kuckelman, Kristin A., Ricky R. Lightfoot, and Debra L. Martin
2002 The Bioarchaeology and Taphonomy of Violence at Castle Rock and Sand
Canyon Pueblos, Southwestern Colorado. American Antiquity 67(3):486-513.
Lambert, Patricia M., Brian R. Billman, and L. Leonard Banks
2000 Explaining Variability in Human Bone Assemblages from the American
Southwest: A Case Study from the Southern Piedmont of Sleeping Ute
Mountain, Colorado. International Journal ofOsteoarchaeology 10:49-64.
Lambert, Patricia M.
1997 Patterns of Violence in Prehistoric Hunter-Gatherer Societies of Coastal
Southern California. In Troubled Times: Violence and Warfare in the Past,
edited by Debra L. Martin and David W. Frayer, pp. 77-110. Gordon and
Breach, Amsterdam.
Larsen, Clark S.
2002 Bioarchaeology: The Lives and Lifestyles of Past People. Journal of
Archaeological Research 10(2):119-166.
125
LeBlanc, Steven A.
1986 Aspects of Southwestern Prehistory: A.D. 900-1400. In Ripples in the
Chichimec Sea: New Considerations ofSouthwestern-Mesoamerican
Interactions, edited by F. J. Mathien and Randall H. McGuire, pp. 105-134.
Southern Illinois University, Carbondale, Illinois.
LeBlanc, Steven A.
1999 Prehistoric Warfare in the American Southwest. University of Utah Press,
Salt Lake City.
Lekson, Stephen H.
1999a The Chaco Meridian: Centers ofPolitical Power in the Ancient Southwest.
Alta Mira, Walnut Creek, California.
1999b Was Casas a Pueblo? In The Casas Grandes World, edited by Curtis F.
Schaafsma and Carol L. Riley, pp. 84-92. University of Utah, Salt Lake City.
Lister, Robert H.
1958 Archaeological Excavations in the Northern Sierra Madre Occidental,
Chihuahua and Sonora, Mexico. University of Colorado series in
anthropology. University of Colorado, Boulder, Colorado.
1978 Mesoamerican Influences at Chaco Canyon, New Mexico. In Across the
Chichimec Sea: Papers in Honor ofJ. Charles Kelley, edited by Carroll L.
Riley and Basil C. Hedrick, pp. 233-241. Southern Illinois University,
Carbondale.
Longin, R.
1971 New Method of Collagen Extraction for Radiocarbon Dating. Nature
230:241-242.
Lovejoy, C. Owen, Richard S. Meindl;T. R. Pryzbeck, and Robert P. Mensforth
1985 Chronological Metamorphosis of the Auricular Surface of the Ilium: A New
Method for the Determination of Adult Skeletal Age at Death. American
Journal ofPhysical Anthropology 68: 15-28.
Lovell, Nancy C.
1997 Trauma Analysis in Paleopathology. American Journal ofPhysical
Anthropology 40:139-170.
Mallet, James
2006 G-Test. Galton Laboratory Software. http://www.ucl.ac.uk/taxome/jim/bin/
software.html. Accessed December 15, 2008.
126
McBrinn, Maxine E. and Laurie D. Webster
2008 Creating an Archaeology Without Borders. In Archaeology Without
Borders: Contact, Commerce and Change in the Us. Southwest and
Northwestern Mexico, edited by Laurie D. Webster, Maxine E. McBrinn, and
Eduardo Gamboa Carrera, pp. 1-24. University of Colorado, Boulder.
McGuire, Randall H.
1980 The Mesoamerican Connection in the Southwest. The Kiva 46(1-2):3-38.
Meindl, Richard S. and C. Owen Lovejoy
1985 Ectocranial Suture Closure: A Revised Method for the Determination of
Skeletal Age at Death Based on the Lateral-Anterior Sutures. American
Journal ofPhysical Anthropology 68:57-66.
Meindl, Richard S., and Katherine F. Russell
1998 Recent Advances in Method and Theory in Paleodemography. Annual
Review ofAnthropology 27:375-399.
Mendiola Galvan, Francisco
2008 Imaginary Border, Profound Border: Terminological and Conceptual
Construction ofthe Archaeology ofNorthern Mexico. In Archaeology Without
Borders: Contact, Commerce, and Change in the Us. Southwest and
Northwestern Mexico, edited by Laurie D. Webster, Maxine E. McBrinn, and
Eduardo Gamboa Carrera, pp. 291-300. University of Colorado, Boulder.
Nelson, Ben A., J. Andrew Darling, and David A Kice
1992 Mortuary Practices and the Social Order at La Quemada, Zacatecas, Mexico.
Latin American Antiquity 3:298-315.
Neves, Walter A, AM. Barros, and M.A Costa
1999 Incidence and Distribution ofPostcranial Fractures in the Prehistoric
Population of San Pedro de Atacama, Northern Chile. American Journal of
Physical Anthropology 109:253-258.
Noe-Nygaard, Nanna
1977 Butchering and Marrow Fracturing as a Taphonomic Factor in
Archaeological Deposits. Paleobiology 3:218-237.
Novak, Shannon A, and Dana D. Kollmann
2000 Perimortem Processing of Human Remains Among the Great Basin
Fremont. International Journal ofOsteoarchaeology 10:65-75.
127
Ogilvie, Marsha D., and Charles E. Hilton
2000 Ritualized Violence in the Prehistoric American Southwest. International
Journal ofOsteoarchaeology 10:27-48.
Ortner, Donald J, and Walter G. J. Putschar
1981 Identification ofPathological Conditions in the Human Skeleton.
Smithsonian Contributions to Anthropology Smithsonian Institution,
Washington, D.C.
Owsley, Douglas W.
1994 Warfare in Coalescent Tradition Populations of the Northern Plains. In
Skeletal Biology in the Great Plains: Migration, Warfare, Health, and
Subsistence, edited by Douglas W. Owsley and Richard L. Jantz, pp. 333-343.
Smithsonian Institution, Washington, D.C.
Owsley, Douglas W., Hugh E. Berryman, and William M. Bass
1977 Demographic and Osteological Evidence for Warfare at the Larson Site,
South Dakota. Plains Anthropologist 22(78): 119-131.
Phenice, T. W.
1969 A Newly Developed Visual Method of Sexing the Os Pubis. American
Journal ofPhysical Anthropology 30:297-302.
Pickering, T.R., M. Dominguez-Rodrigo, C.P. Egeland, and C.K. Brain
2005 The Contribution of Limb Bone Fracture Patterns to Reconstructing Early
Hominid Behavior at Swartkrans Cave (South Africa): Archaeological
Application of a New Analytical Method. International Journal of
Osteoarchaeology 15:247-260.
Pijoan, Carmen M., and Josefina Mansilla
1990 Evidencias Rituales en Restos Humanos del Norte de Mesoamerica. In
Mesoamerica y Norte de Mexico: Siglo lX-Xll, edited by Federica Sodi
Miranda, pp. 467-478. Museo Nacional de Antropologia, Mexico City.
Pijoan, C. M., J. Mansilla, 1. Leboreiro, V. H. Lara, and P. Bosch
2007 Thermal Alterations in Archaeological Bones. Archaeometry 49:713-727.
Price, T. Douglas, and Gary M. Feinman
1993 Images ofthe Past. Mayfield, Mountain View, California.
Ramsey, Fred L., and Daniel W. Schafer
1997 The Statistical Sleuth: A Course in Methods ofData Analysis. Duxbury,
Belmont, California.
128
Rathbun, Ted A., and Jane E. Buikstra
1984 Human Identification: Case Studies in Forensic Anthropology. Charles C.
Thomas, Springfield, Illinois.
Ravesloot, John C.
1988 Mortuary Practices and Social Differentiation at Casas Grandes,
Chihuahua, Mexico. Anthropological Papers of the University of Arizona
University ofArizona, Tucson.
Ravesloot, John C.
1984 Social Differentiation at Casas Grandes, Chihuahua, Mexico: An
Archaeological Analysis ofMortuary Practices. Southern Illinois University.
Redfern, Rebecca
2008 New Evidence for Iron Age Secondary Burial Practice and Bone
Modification from Gussage All Saints and Maiden Castle (Dorset, England).
Oxford Journal ofArchaeology 27:281-302.
Reimer, P. J., M.G.L. Baillie, E. Bard, A. Bayliss, J.W. Beck, C.J.H. Bertrand, P.G.
Blackwell, C.E. Buck, G.S. Burr, K.B. Cutler, P.E. Damon, R.L. Edwards, R.G.
Fairbanks, M. Friedrich, T.P. Guilderson, A.G. Hogg, K.A. Hughen, B. Kromer, F.G.
McCormac, S.W. Manning, C.B. Ramsey, RW. Reimer, S. Remmele, lR Southon,
M. Stuiver, S. Talamo, F.W. Taylor, J. van der Plicht, and C.E. Weyhenmeyer
2004 IntCal04 Terrestrial Radiocarbon Age Calibration, 26 - 0 ka BP.
Radiocarbon 46: 1029-1 058.
Riley, Carroll L.
2005 Becoming Aztlan: Mesoamerican Influence in the Greater Southwest A.D.
1200-1500. University ofUtah, Salt Lake City.
1993 Charles C. DiPeso: An Intellectual Biography. In Culture and Contact:
Charles C. DiPeso's Gran Chichimeca, edited by Anne 1. Woosley and John
C. Ravesloot, pp. 11-22. Amerind Foundation, Dragoon, Arizona.
Roksandic M.
2002 Position of Skeletal Remains as a Key to Understanding Mortuary Behavior.
In Advances in Forensic Taphonomy. Method, Theory, and Archaeological
Perspectives, edited by William D. Haglund and Marcella H. Sorg, pp. 100-
117. C.RC. Press, London.
Roksandic, M., M. Djuric, Z. Rakosevic, and K. Seguin
2006 Interpersonal Violence at Lepenski Vir MesolithiclNeolithic Complex of
the Iron Gates Gorge (Serbia-Romania). American Journal ofPhysical
Anthropology 129:339-348.
129
Sauer, Nonnan J.
1998 The Timing of Injuries and Manner of Death: Distinguishing Among
Antemortem, Perimortem, and Postmortem Trauma. In Forensic Osteology:
Advances in the Identification ofHuman Remains, edited by Kathleen J.
Reichs, pp. 321-332. C.C. Thomas, Springfield, Illinois.
Saul, Frank P.
1972 The Human Skeletal Remains ofAltar de Sacrificios: An Osteobiographic
Analysis. Papers of the Peabody Museum Peabody Museum, Cambridge,
Massachusetts.
Schaafsma, Curtis F., and Carroll L. Riley
1999 Introduction. In The Casas Grandes World, pp. 3-11. University of Utah,
Salt Lake City.
Schroeder, S.
2001 Secondary Disposal of the Dead: Cross Cultural Codes. World Cultures
12:77-93.
Schurr, Mark R.
1989 Fluoride Dating of Prehistoric Bones by Ion Selective Electrode. Journal of
Archaeological Science 16:265-270.
2007 University of Notre Dame Fluoride Dating Service Center Homepage.
http://www.nd.eduJ~mschurrlfluoride/. Accessed December 10,2008.
Scott, E.C.
1979 Dental Wear Scoring Technique. American Journal ofPhysical
Anthropology 51 :213-218.
Shipman, Pat, and Jennie Rose
1983 Evidence of Butchery and Hominid Activities at Torralba and Ambrona: An
Evaluation Using Microscopic Techniques. Journal ofArchaeological Science
10:465-474.
Smith, B.H.
1984 Patterns of Molar Wear in Hunter-Gatherers and Agriculturalists. American
Journal ofPhysical Anthropology 63 :39-56.
Smith, M. O.
1996 'Parry' Fractures and Female-Directed Interpersonal Violence: Implications
from the Late Archaic Period of West Tennessee. International Journal of
Osteoarchaeology 6:84-91.
130
Sokal, Robert R., and F. James Rohlf
1994 Biometry: The Principles and Practice ofStatistics in Biological Research.
W.H. Freeman, San Francisco.
Spencer, Frank
1990 Piltdown: A Scientific Forgery. Oxford University, Oxford.
Stuiver, M., and P.J. Reimer
1993 Extended 14C Database and Revised CALIB Radiocarbon Calibration
Program. Radiocarbon 35:215-230.
Stuiver, M., and P.J. Reimer
2005 CALIB 5.0., http://calib.qub.ac.uk/calib/. Accessed February 17,2009.
Sugiyama, Saburo
2005 Human Sacrifice, Militarism, and Ruleship: Materialization ofState
Ideology at the Feathered Serpent Pyramid, Teotihuacan. Cambridge
University, Cambridge, UK.
Sutherland, Leslie D., and Judy Myers Suchey
1991 Use of the Ventral Arc in Pubic Sex Determination. Journal ofForensic
Sciences 36(2):501-511.
Swanson, Steven
2003 Documenting Prehistoric Communication Networks: A Case Study in the
Paquime Polity. American Antiquity 68:753-767.
Taylor, Royal E.
1997 Radiocarbon Dating. In Chronometric Dating in Archaeology, edited by
Royal E. Taylor and Martin J. Aitken, pp. 60-92. Plenum, New York.
Tiesler, Vera, and Andrea Cucina
2006 Procedures in Human Heart Extraction and Ritual Meaning: A Taphonomic
Assessment of Anthropogenic Marks in Classic Maya Skeletons. Latin
American Antiquity 17(4):493-510.
Torres-Rouff, Christina, and Maria Antonietta Costa Junqueira
2006 Interpersonal Violence in Prehistoric San Pedro de Atacama, Chile:
Behavioral Implications of Environmental Stress. American Journal of
Physical Anthropology 130:60-70.
131
Tung, Tiffiny A.
2007 Trauma and Violence in the Wari Empire of the Peruvian Andes: Warfare,
Raids, and Ritual Fights. American Journal ofPhysical Anthropology
133:941-956.
Turner II, Christy G., and Nancy T. Morris
1970 A Massacre at Hopi. American Antiquity 35(3):320-331.
Turner II, Christy G., and Jacqueline A. Turner
1999 Man Corn: Cannibalism and Violence in the Prehistoric American
Southwest. University of Utah, Salt Lake City.
Ubelaker, Douglas H.
1989 Human Skeletal Remains: Excavation, Analysis, Interpretation. Taraxacum,
Washington, D.C.
VanPool, Christine S., and Todd L. VanPool
2007 Signs ofthe Casas Grandes Shamans. University of Utah, Salt Lake City.
Villa, Paola, Claude Bouville, Jean Courtin, Daniel Helmer, Eric Mahieu, Pat
Shipman, Giorgio Belluomini, and Marili Branca
1986 Cannibalism in the Neolithic. Science 233(4762):431-437.
Walker, Cameron M.
2006 Bioarchaeology ofNewly Discovered Burial Caves in the Sierra
Tarahumara. Ph.D. dissertation, University of Oregon, Eugene. University
Microfilms, Ann Arbor.
Walker, Philip L.
2001 A Bioarchaeological Perspective on the History of Violence. Annual
Review ofAnthropology 30:573-596.
1989 Cranial Injuries as Evidence of Violence in Prehistoric Southern California.
American Journal ofPhysical Anthropology 80:313-323.
1997 Wife Beating, Boxing, and Broken Noses: Skeletal Evidence for the
Cultural Patterning of Violence. In Troubled Times: Violence and Warfare in
the Past, edited by Debra L. Martin and David W. Frayer, pp. 145-180.
Gordon and Breach, Amsterdam.
Walker, William H.
2002 Stratigraphy and Practical Reason. American Anthropologist 104:159-177.
132
Wallerstein, Immanuel M.
1974 Capitalist Agriculture and the Origins ofthe European World-Economy in
the Sixteenth Century. Academic Press, New York.
Weaver, David S.
1981 An Osteological Test of Changes in Subsistence and Settlement Patterns at
Casas Grandes, Chihuahua, Mexico. American Antiquity 46:361-364.
Webster, Laurie D., Maxine E. McBrinn, and Eduardo Gamboa Carrera
2008 Archaeology Without Borders: Contact, Commerce, and Change in the us.
Southwest and Northwestern Mexico. University of Colorado, Boulder.
Weiss-Krejci, Estella
2001 Restless Corpses: 'Secondary Burial' in the Babenberg and Habsburg
Dynasties. Antiquity 75:769-780.
Whalen, Michael E., and Paul E. Minnis
1996 Ballcourts and Regional Organization in the Casas Grandes Region.
American Antiquity 61 :732-746.
2001a Casas Grandes and Its Hinterland: Prehistoric Regional Organization in
Northwest Mexico. University of Arizona, Tucson.
2001 b The Casas Grandes Regional System: A Late Prehistoric Polity of
Northwestern Mexico. Journal ofWorld Prehistory 15(3):313-364.
2003 The Local and the Distant in the Origin of Casas Grandes, Chihuahua,
Mexico. American Antiquity 68:314-332.
White, Tim D.
1992 Prehistoric Cannibalism at Mancos 5MTUMR-2346. Princeton University
Press, Princeton.
White, Tim D., and Pieter A. Folkens
2005 The Human Bone Manual. Elsevier Acadamic Press, Amsterdam.
Wilcox, David R.
1991 The Mesoamerican Ballgame in the American Southwest. In The
Mesoamerican Ballgame, edited by Vernon L. Scarborough and David R.
Wilcox, pp. 101-128. University of Arizona, Tucson.
133
Wilkerson, S. Jeffrey K.
1991 And Then They Were Sacrificed: The Ritual Ballgame ofNortheastern
Mesoamerica Through Time and Space. In The Mesoamerican Ballgame,
edited by Vernon L. Scarborough and David R. Wilcox, pp. 45-72. University
ofArizona, Tucson.
Willey, P.
1990 Prehistoric Wwfare on the Great Plains: Skeletal Analysis ofthe Crow
Creek Massacre Victims. Garland, New York.
Woodall, J. Ned
1968 Growth Arrest Lines in Longbones of the Casas Grandes Population. Plains
Anthropologist 13(40): 152-160.
Woosley, Ann 1., and Bart Olinger
1993 The Casas Grandes Ceramic Tradition: Production and Regional Exchange
of Ramos Polychrome. In Culture and Contact: Charles DiPeso's Grand
Chichimeca, edited by Ann 1. Woosley and John C. Ravesloot, pp. 105-132.
University ofNew Mexico, Albuquerque.